System

Abstract

To provide a system.SOLUTION: A system comprises: means for automatically answering questions from investors; means for presenting summaries of financial reports and short reports, and points of concern; means for answering indicators other than the financial reports and short reports; means for answering the investors' questions; means for recognizing the investors' emotions with the usage of an emotion engine to present information according to the emotions; and means for extracting important information from the financial reports and short reports with the usage of text mining technology.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 a description and related instruction sentence regarding 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] JP 2022-180282 A Summary of the Invention [Problem to be solved by the invention]

[0004] In order for investors to quickly and accurately obtain the information they need, they need to understand the contents of financial reports and short messages and extract the important information contained therein. However, this information requires specialized knowledge and is huge in volume, making it difficult for investors to understand everything. Furthermore, when investors have questions, they have to go to the trouble of contacting experts to get answers. [Means for solving the problem]

[0005] The present invention provides a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering KPIs other than financial reports, and a means for answering investors' questions. This allows investors to obtain information quickly and accurately, reducing the effort required for collecting investment information. In addition, by using natural language processing technology and text mining technology, it is possible to automate the analysis and extraction of information that requires specialized knowledge. [Brief description of the drawings]

[0006] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Diagram 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. FIG. [Diagram 3] FIG. 11 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Diagram 5] FIG. 13 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. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 13 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. 2 is a sequence diagram showing a process flow of the data processing system in the first embodiment of the first form example. [Figure 12]FIG. 11 is a sequence diagram showing a process flow of the data processing system in the application example 1 of the first embodiment. [Figure 13] FIG. 11 is a sequence diagram showing a process flow of the data processing system in the second embodiment of the second embodiment. [Figure 14] FIG. 13 is a sequence diagram showing a process flow of the data processing system in application example 2 of embodiment 2. [Figure 15] FIG. 13 is a sequence diagram showing a process flow of a data processing system in a third embodiment of the present invention. [Figure 16] FIG. 13 is a sequence diagram showing a process flow of the data processing system in the third application example of the third embodiment. [Figure 17] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Form Example 1 when combined with an emotion engine. [Figure 19] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system according to the embodiment 2 of the second embodiment in the case where an emotion engine is combined. [Figure 20] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Form Example 2 when combined with an emotion engine. [Figure 21] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in the embodiment 3 of the form example 3 in the case where an emotion engine is combined. [Figure 22] FIG. 13 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Form Example 3 when combined with an emotion engine. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0007] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0008] First, the terms used in the following description will be explained.

[0009] In the following embodiments, a signed processor (hereinafter simply referred to as a "processor") may be one arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be one 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), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0010] In the following embodiments, a signed RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by the processor.

[0011] 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.

[0012] In the following embodiments, a communication I / F (Interface) with a code is an interface including a communication processor and an antenna. The communication I / F controls communication between multiple computers. An example of a communication standard applied to the communication I / F is a wireless communication standard including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0013] 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. In addition, in this specification, the same idea as "A and / or B" is also applied when three or more things are expressed by connecting them with "and / or."

[0014] [First embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] 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.

[0017] 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 wide area network (WAN) and / or a local area network (LAN).

[0018] 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.

[0019] The reception device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A detects contact with a pointer (e.g., a pen or a finger) to receive user input by the touch of the pointer. 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.

[0020] 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 (e.g., voice and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs voice according to instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a Complementary Metal-Oxide-Semiconductor (CMOS) image sensor or a Charge Coupled Device (CCD) image sensor.

[0021] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54.

[0022] FIG. 2 shows an example of main functions of the data processing device 12 and the smart device 14.

[0023] As shown in Fig. 2, 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. The specific process program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific process program 56 from the storage 32, and executes the read specific process 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 process program 56 executed on the RAM 30.

[0024] 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.

[0025] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores a reception output program 60. The reception output program 60 is used together with the specific processing program 56 by the data processing system 10. The processor 46 reads out the reception output program 60 from the storage 50, and executes the read reception output program 60 on the RAM 48. The reception output processing 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.

[0026] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0027] "Example 1"

[0028] One embodiment of the present invention is a chatbot system for investors. The system receives questions from investors and analyzes the content of the questions using natural language processing technology. Based on the analysis results, the system generates an appropriate answer and provides it to the investor. For example, if an investor asks, "What are the main points of the latest financial report?", the system analyzes the financial report materials, extracts the main points, and answers the investor.

[0029] "Example 2"

[0030] Another embodiment of the present invention is a function for presenting summaries of financial report materials and short messages and points of concern. This function uses text mining technology to extract important information from financial report materials and short messages and present it to investors. For example, important indicators such as sales and profit margins listed in financial report materials are extracted and summarized information is provided to investors.

[0031] "Example 3"

[0032] Furthermore, another embodiment of the present invention has a function for answering KPIs other than financial reports. This function provides answers to questions from investors about KPIs that are not listed in financial reports. For example, if an investor asks, "What is the level of customer satisfaction?", the system obtains information from sources other than financial reports, such as the results of a customer satisfaction survey, and answers the investor.

[0033] The process flow of each embodiment will be described below.

[0034] "Example 1"

[0035] Step 1: The system receives the investor's question.

[0036] Step 2: The system uses natural language processing techniques to analyze the content of the question.

[0037] Step 3: Based on the analysis results, the system generates an appropriate answer.

[0038] Step 4: Provide the generated responses to investors.

[0039] "Example 2"

[0040] Step 1: The system retrieves financial report materials and short reports.

[0041] Step 2: Use text mining techniques to extract important information from the acquired materials.

[0042] Step 3: Summarize the extracted information and present it to investors.

[0043] "Example 3"

[0044] Step 1: The system receives the investor's question.

[0045] Step 2: The system obtains the required information from sources other than financial statements.

[0046] Step 3: Based on the information obtained, the system answers the investor's questions.

[0047] Example 1

[0048] Next, a description will be given of Example 1 of the embodiment 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".

[0049] Conventional investor relations systems had the problem that it was difficult to quickly and accurately respond to questions from investors. In addition, there was a lack of a way to efficiently extract information on financial reports and other indicators and provide it to investors. This meant that investors had to spend a lot of time and effort to obtain the information they needed.

[0050] 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.

[0051] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting a summary of financial report materials and important points, a means for answering indicators other than financial reports, a means for analyzing questions using natural language processing technology and generating appropriate answers, a means for generating answers using a generative AI model, a means for generating prompt sentences and inputting them into the generative AI model, and a means for providing the generated answers to investors, thereby enabling investors to quickly and accurately obtain necessary information.

[0052] "Investor" refers to an individual or entity that invests money in stocks, bonds, and other financial instruments in the financial markets.

[0053] "Means of automatically answering questions" refers to the function of receiving questions from investors and having the system automatically generate and provide appropriate answers.

[0054] "Financial report materials" refers to reports on financial status and performance that are issued periodically by a company.

[0055] "Means of presenting summaries and key points" refers to the function of extracting important information from financial report documents, summarizing it concisely and presenting it to investors.

[0056] An "indicator" refers to numerical data or statistical information used to evaluate a company's performance and financial situation.

[0057] "Natural language processing technology" refers to technology that enables computers to understand and analyze human language.

[0058] A "generative AI model" refers to a model for generating text data using artificial intelligence.

[0059] A "prompt sentence" refers to an instruction sentence to be input into a generative AI model.

[0060] "Analysis" refers to the process of breaking down input data and understanding its meaning and structure.

[0061] "Means for generating an answer" refers to the function of creating an appropriate answer based on the analysis results.

[0062] "Means of providing" refers to the functionality for communicating the generated answers to investors.

[0063] This invention relates to a chatbot system for investors, which provides quick and accurate answers to questions from investors. This system operates among three parties: a server, a terminal, and a user.

[0064] The server receives questions from investors and analyzes them using natural language processing technology. For this analysis, a generative AI model such as OpenAI's GPT-4 is used. The server extracts the intent and keywords of the question and obtains information to generate an appropriate answer.

[0065] The server generates an appropriate answer based on the analysis results. For example, the server refers to the latest financial report documents stored in the database and extracts the main points. In this case, the server inputs the following prompt sentences to the generative AI model:

[0066] Prompt: "What are the key takeaways from your latest earnings report?"

[0067] The generative AI model will generate an answer based on this prompt, for example in the following format:

[0068] Answer: "The main takeaways from our most recent financial report were that revenue increased 10% year over year and net income increased 15%. We also saw strong sales of our new products, which gives us good prospects for future growth."

[0069] The server transmits the generated answer to the terminal. The terminal receives the answer transmitted from the server and displays it to the user. The user can check the answer on the screen of the terminal.

[0070] The specific hardware configuration of this system is that the server is a computer equipped with a high-performance processor and large-capacity memory, and the terminal is a device operated by the user, such as a PC or smartphone. The software includes a natural language processing engine, generative AI model, and database management system that run on the server side.

[0071] As a concrete example, consider the case where a user inputs a question such as "What are the main points of the latest financial report?" The device sends this question to the server, which analyzes it using GPT-4. The server retrieves the latest financial report materials from the database and extracts the main points. The server inputs the prompt sentence into the generative AI model and sends the generated answer to the device. The device displays the answer to the user, allowing the user to quickly obtain the information they need.

[0072] Thus, the present invention provides a system that allows investors to quickly and accurately obtain the information they need.

[0073] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0074] Step 1:

[0075] The user enters a question.

[0076] A user inputs a question about an investment into a terminal (PC or smartphone). For example, the user inputs, "What are the main points of the latest financial report?" The input question is stored as text data on the terminal.

[0077] Step 2:

[0078] The terminal sends a question to the server.

[0079] The terminal transmits the question input by the user to the server. At this time, the terminal transmits text data to the server via the Internet. The input data is the question text of the user.

[0080] Step 3:

[0081] The server receives the query.

[0082] The server receives the question sent from the terminal. The received question is stored in the server as text data. The input data is the user's question text sent from the terminal.

[0083] Step 4:

[0084] The server parses the question.

[0085] The server analyzes the received question using natural language processing technology. For this analysis, it uses a generative AI model such as OpenAI's GPT-4. The server extracts the intent and keywords of the question and obtains information to generate an appropriate answer. The input data is the user's question text, and the output data is the analysis result.

[0086] Step 5:

[0087] The server generates an appropriate response.

[0088] The server generates an appropriate answer based on the analysis results. For example, the server refers to the latest financial report documents stored in the database and extracts the main points. In this case, the server inputs the following prompt sentences to the generative AI model:

[0089] Prompt: "What are the key takeaways from your latest earnings report?"

[0090] The generative AI model generates an answer based on this prompt. The input data is the prompt and the analysis result, and the output data is the generated answer.

[0091] Step 6:

[0092] The server sends the response to the terminal.

[0093] The server transmits the generated answer to the terminal. At this time, the server transmits the answer to the terminal as text data. The input data is the generated answer text, and the output data is the answer text transmitted to the terminal.

[0094] Step 7:

[0095] The terminal displays the answer to the user.

[0096] The terminal receives the answer sent from the server and displays it to the user. The user can check the answer on the screen of the terminal. The input data is the answer text sent from the server, and the output data is the answer text displayed to the user.

[0097] (Application example 1)

[0098] Next, a description will be given of Application Example 1 of Form 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."

[0099] Conventional chatbot systems for investors have the ability to automatically answer questions from investors, but they lack the ability to provide appropriate answers to security-related questions.In addition, they have issues with not improving the user experience because they do not provide a user interface or retrieve information from a database efficiently.

[0100] 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.

[0101] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering performance indicators other than financial reports, a means for answering investors' concerns, a means for generating appropriate answers to security-related questions, a means for analyzing questions and generating appropriate answers using natural language processing technology, a means for acquiring information from a database and providing it to users, and a means for providing a user interface. This makes it possible to provide quick and accurate answers to questions from investors and provide comprehensive support including security-related information.

[0102] "Means of automatically responding to questions from investors" refers to a function that analyzes questions entered by investors and automatically generates and provides appropriate answers.

[0103] The "means of presenting summaries of financial report documents and short messages as well as points of interest" is a function that extracts important information from financial report documents and short messages and presents the user with summaries and points of particular interest.

[0104] The "means of answering questions about performance indicators other than financial statements" is a function that provides appropriate answers to questions about performance indicators other than financial statements.

[0105] "Means of answering investor questions" is a function that provides appropriate answers to questions that investors have.

[0106] The "means for generating appropriate answers to security-related questions" is a function for analyzing security questions and generating and providing appropriate answers.

[0107] "Means of analyzing questions using natural language processing technology and generating appropriate answers" refers to a function that uses natural language processing technology to analyze a user's question and generate an appropriate answer based on the content of the question.

[0108] "Means for retrieving information from a database and providing it to a user" refers to a function for searching information stored in a database, retrieving the necessary information, and providing it to a user.

[0109] "Means for providing a user interface" refers to a function that provides an interface for a user to input a question and receive a response.

[0110] A system for implementing this invention includes means for automatically answering questions from investors, means for presenting summaries of financial report materials and short messages and points of concern, means for answering performance indicators other than financial reports, means for answering investor concerns, means for generating appropriate answers to security-related questions, means for analyzing questions and generating appropriate answers using natural language processing technology, means for retrieving information from a database and providing it to a user, and means for providing a user interface.

[0111] System Configuration

[0112] The server includes a question analysis engine using natural language processing technology, a database access module, an answer generation engine, and a user interface module. These modules are implemented using Python and natural language processing libraries (e.g., spaCy and Transformers).

[0113] Program processing explanation

[0114] When the server receives a question from a user, it first analyzes the question using natural language processing technology. Natural language processing libraries such as spaCy and Transformers are used for this analysis. The analyzed question is used to retrieve related information from a database via a database access module. The database contains financial report documents, performance indicators, security-related information, and so on.

[0115] The acquired information is processed into appropriate answers by the answer generation engine, which uses a generative AI model to generate answers, and the generated answers are provided to the user through a user interface module.

[0116] Examples

[0117] For example, if a user asks, "What are the latest security threats?" the system would do the following:

[0118] 1. The user interface module receives a question.

[0119] 2. Analyze the question using natural language processing techniques.

[0120] 3. The database access module retrieves the relevant security information from the database.

[0121] 4. The answer generation engine generates appropriate answers based on the information obtained.

[0122] 5. The user interface module provides the generated answers to the user.

[0123] Examples of prompt statements

[0124] User: What are the latest security threats?

[0125] Systems: The latest security threat is ransomware attacks.

[0126] In this way, the system can provide quick and accurate answers to questions from investors and users, providing comprehensive support, including security-related information.

[0127] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0128] Step 1:

[0129] The user inputs a question. The user inputs a question through the user interface of a smartphone or a robot. The input question is sent to the server.

[0130] Step 2:

[0131] The server receives the question. The server receives the question sent by the user and analyzes the question using natural language processing technology. Specifically, it uses natural language processing libraries such as spaCy and Transformers to analyze the meaning of the question. The input is the user's question, and the output is the structure of the analyzed question.

[0132] Step 3:

[0133] The server retrieves information from the database. Based on the content of the parsed question, the server searches the database for relevant information. The database may contain financial report documents, performance indicators, security-related information, etc. The input is the parsed question structure, and the output is the relevant information.

[0134] Step 4:

[0135] The server generates an answer. The server uses a generative AI model to generate an appropriate answer based on the acquired information. Specifically, the acquired information is input as a prompt into the generative AI model to generate an appropriate answer. The input is the information acquired from the database, and the output is the generated answer.

[0136] Step 5:

[0137] The server provides the answer to the user. The server provides the generated answer to the user through a user interface module. The user can check the answer on the screen of a smartphone or a robot. The input is the generated answer, and the output is the answer provided to the user.

[0138] Specifically, when a user asks "What are the latest security threats?", the server receives the question, analyzes it using natural language processing technology, and retrieves information about the latest security threats from a database. It then uses a generative AI model to generate the answer "The latest security threat is a ransomware attack" and provides it to the user.

[0139] Example 2

[0140] Next, a second embodiment of the second embodiment will be described. 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."

[0141] In the conventional system, investors had to manually extract and summarize important information from financial reports and short messages, which was cumbersome and time-consuming. It was also difficult to quickly and appropriately respond to questions from investors, so an efficient means of resolving investors' concerns was needed.

[0142] 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.

[0143] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering questions about indicators other than financial reports, a means for uploading financial report materials and short messages, a means for receiving uploaded materials, a means for analyzing materials using a text mining tool, a means for extracting important indicators, a means for summarizing the extracted information, a means for formatting the summary in natural language using a generative AI model, a means for transmitting the formatted summary information to a terminal, and a means for the terminal to present the summary information to a user. This makes it possible to quickly and accurately provide information required by investors and efficiently resolve investors' questions.

[0144] An "investor" is an individual or legal entity that provides capital to a company or project and expects to receive a profit in return.

[0145] "Means for automatically answering questions" refers to a part of a system that has the function of receiving questions from investors, analyzing them using natural language processing technology, and generating appropriate answers.

[0146] A "financial report" is a document prepared by a company at regular intervals to report on its financial status and business performance.

[0147] A "tanshin" is a report that briefly summarizes a company's financial results.

[0148] A summary is a short sentence that concisely summarizes a longer piece of text or complex information.

[0149] "Means of presenting sections of points of concern" refers to part of a system that has the function of extracting and presenting important information and sections of interest to investors from financial report documents and short messages.

[0150] An "indicator" is a number or data used to evaluate a company's performance and financial situation.

[0151] An "uploading means" is a part of the system that has the functionality to allow a user to send files to the system.

[0152] The "receiving means" is a part of the system in which the server has the function of receiving and storing files sent by users.

[0153] A "text mining tool" is software for analyzing text data and extracting important information.

[0154] "Means for analysis" refers to a part of a system that has the function of analyzing the contents of documents using text mining tools.

[0155] The "means for extracting important indicators" is a part of the system that has the function of extracting important data such as sales revenue and profit margins from the analysis results.

[0156] The "summarizing means" is the part of the system that has the functionality to generate a concise summary based on the extracted key information.

[0157] A "generative AI model" is a model for generating sentences in natural language using artificial intelligence technology.

[0158] A "means for formatting a summary in natural language" is a part of a system that has the function of formatting summary information in easy-to-understand natural language using a generative AI model.

[0159] A "terminal" is a device through which a user accesses the system and receives information.

[0160] A "means for presenting" is a part of the system where a terminal has the ability to display information to a user.

[0161] This invention is a system that allows investors to quickly and accurately extract important information from financial reports and short messages, and provides summary information. The system operates in cooperation with a server, a terminal, and a user.

[0162] Hardware and Software Used

[0163] Hardware: Servers, terminals

[0164] Software: Text mining tools (e.g. NLTK, SpaCy), generative AI models (e.g. GPT-4)

[0165] System Operation Overview

[0166] 1. The user uploads financial report documents and short messages to the system.

[0167] 2. The server receives the uploaded materials and analyzes the text data using text mining tools.

[0168] 3. The server extracts key indicators such as sales revenue and profit margins from the analysis results.

[0169] 4. The server summarizes the extracted information and uses a generative AI model to format it in natural language for investors to understand.

[0170] 5. The server sends the formatted summary information to the terminal and presents it to the user.

[0171] Examples

[0172] For example, if a user uploads "Financial report materials for Q1 2023" to the system, the server will process it as follows:

[0173] 1. The user uploads the "2023 Q1 Financial Report Materials" to the system.

[0174] 2. The server receives the material and analyzes the text data using NLTK.

[0175] 3. The server extracts key indicators such as sales revenue and profit margins.

[0176] 4. The server summarizes the extracted information and uses GPT-4 to generate a natural language summary such as, "Revenue in Q1 2023 increased 10% compared to the same period last year, and profit margins reached 15%."

[0177] 5. The server sends the generated summary information to the terminal and presents it to the user.

[0178] Examples of prompt statements

[0179] An example of a prompt to be input to the generative AI model is as follows:

[0180] "Please extract and summarize key indicators from the text below. Text: 'In the financial results report for Q1 2023, sales increased 10% year-on-year and profit margins reached 15%. In addition, sales of new products are strong, and future growth is expected.'"

[0181] Using this prompt, the generative AI model extracts key information and generates a summary.

[0182] This system allows investors to quickly obtain the information they need, enabling them to make efficient investment decisions. It also allows for quick and appropriate responses to investor questions, resolving investors' doubts and providing highly reliable information.

[0183] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0184] Step 1:

[0185] The user uploads financial report documents and short messages to the system.

[0186] Specific operation: The user selects "Financial report materials for Q1 2023.pdf" and clicks the upload button.

[0187] Input: The financial report file selected by the user.

[0188] Output: The uploaded file is sent to the server.

[0189] Step 2:

[0190] The server receives the uploaded material.

[0191] Specific operation: The server receives "2023Q1Financial Report Materials.pdf" and saves it in the / uploads directory.

[0192] Input: Financial report file submitted by user.

[0193] Output: The file saved in the specified directory on the server.

[0194] Step 3:

[0195] The server analyzes the material using text mining tools.

[0196] Specific operation: The server uses NLTK to extract text from "2023Q1Financial Report Materials.pdf" and begins analysis.

[0197] Input: Financial report file stored on the server.

[0198] Output: The extracted text data.

[0199] Step 4:

[0200] The server extracts key indicators.

[0201] Specific operation: The server extracts important indicators based on keywords such as "sales revenue," "profit margin," and "year-on-year comparison."

[0202] Input: The extracted text data.

[0203] Output: Extracted key indicators data.

[0204] Step 5:

[0205] The server summarizes the extracted information.

[0206] Specific operation: The server generates summary information based on the extracted important indicators.

[0207] Input: Extracted key indicators data.

[0208] Output: Summarized information.

[0209] Step 6:

[0210] The server uses a generative AI model to format the summary in natural language.

[0211] How it works: The server inputs a prompt sentence into GPT-4, which generates a natural language summary such as, "Sales in Q1 2023 increased 10% year-on-year, and profit margins reached 15%."

[0212] Input: The summarized information and the prompt.

[0213] Output: Natural language formatted summary information.

[0214] Step 7:

[0215] The server transmits the formatted summary information to the terminal.

[0216] Specific operation: The server sends the generated summary information to the user's terminal as an HTTP response.

[0217] Input: A formatted summary in natural language.

[0218] Output: Summary information sent to the user's terminal.

[0219] Step 8:

[0220] The terminal presents the summary information to the user.

[0221] Specific behavior: The device displays on the browser, "Revenue in the first quarter of 2023 increased 10% year-on-year, and profit margin reached 15%."

[0222] Input: Summary information sent by the server.

[0223] Output: Summary information displayed to the user.

[0224] (Application example 2)

[0225] Next, a description will be given of Application Example 2 of Form 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."

[0226] Conventional investor information systems have difficulty quickly and accurately extracting and summarizing important information from financial reports and short messages. They also lack the ability to automatically generate appropriate answers to questions from investors. This means that investors have to spend a lot of time and effort to obtain the information they need.

[0227] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means. In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering indicators other than the financial report, and a means for extracting important information from the financial report materials and generating summaries through an application installed on a smartphone. This enables investors to quickly and accurately obtain the information they need.

[0228] An "investor" is an individual or legal entity that provides funds to a company or project with the aim of receiving a profit in return.

[0229] "Means for automatically answering questions" refers to technological means for analyzing questions from investors and generating appropriate answers.

[0230] A "financial report" is an official document that reports a company's financial status and performance.

[0231] A "tanshin" is a document that provides a concise report of a company's quarterly performance and financial status.

[0232] A "summary" is a short summary of a longer document or information, extracting and concisely presenting the important points.

[0233] "Means of presenting the sections of information of particular interest to investors" refers to technological means for extracting and presenting information of particular interest to investors from financial report documents and short messages.

[0234] An "indicator" is a specific number or data used to evaluate a company's performance and financial situation.

[0235] A "smartphone" is a mobile phone that can connect to the Internet and install applications.

[0236] An "Application" is a software program designed to provide a particular function or service.

[0237] "Text mining technology" is a data analysis technique for extracting useful information from large amounts of text data.

[0238] "Natural language processing technology" is a technology that enables computers to understand and analyze human language.

[0239] "Material information" refers to key data and indicators contained in financial reports and short reports that investors need to make decisions.

[0240] A "means for generating a summary" is a technical means for extracting important information from a long document and summarizing it concisely.

[0241] A system for implementing the present invention includes a server, a smartphone, and related software components. A specific configuration and operation of the system will be described below.

[0242] System Configuration

[0243] 1. Server:

[0244] The server uses natural language processing and text mining technologies to analyze questions from investors and generate appropriate answers.

[0245] The server extracts important information from financial reports and short messages and generates summaries.

[0246] The server then sends the generated summary and answers to the smartphone.

[0247] 2. Smartphone:

[0248] The smartphone displays the summary and answers sent from the server through a dedicated application.

[0249] A user can use a smartphone to input a question and send it to the server.

[0250] 3. Software Components:

[0251] A natural language processing library (e.g. spaCy): Used to parse investor questions and generate appropriate answers.

[0252] Text mining libraries (e.g. scikit-learn): Used to extract key information from financial reports and short reports.

[0253] Data processing and calculation

[0254] 1. Question analysis and answer generation:

[0255] The server receives questions from investors and analyzes them using natural language processing techniques.

[0256] Based on the parsed question, an appropriate answer is generated.

[0257] The generated answers are sent to the smartphone and displayed to the user.

[0258] 2. Generate summary of financial report documents:

[0259] The server receives financial report documents and short messages and uses text mining technology to extract important information.

[0260] A summary is generated based on the extracted information.

[0261] The generated summary is sent to the smartphone and displayed to the user.

[0262] Examples

[0263] Text from the financial report: "Financial report for fiscal year 2023. Sales increased 10% year-on-year, and profit margins improved 5%. Sales of new products were strong, with particularly notable growth in Asian markets."

[0264] Example prompt: "Please extract and summarise the key information from this financial statement."

[0265] This system allows investors to quickly and accurately obtain the information they need, improving the efficiency of their decision-making. In addition, by using a smartphone, investors can enjoy the convenience of being able to access information anytime, anywhere.

[0266] The flow of the specific process in the application example 2 will be described with reference to FIG.

[0267] Step 1:

[0268] A user inputs a question using a smartphone and sends it to the server. The input question is sent to the server as text data.

[0269] Step 2:

[0270] The server analyzes the received question using natural language processing technology (e.g., spaCy). Specifically, it analyzes the grammatical structure of the question and extracts important keywords and phrases. The input is the user's question text, and the output is the analyzed keywords and phrases.

[0271] Step 3:

[0272] The server generates appropriate answers based on the analyzed keywords and phrases. It uses a generative AI model to generate the best answer to the question. The input is the analyzed keywords and phrases, and the output is the generated answer text.

[0273] Step 4:

[0274] The server sends the generated answer to the smartphone, which displays the received answer to the user. The input is the generated answer text, and the output is the answer displayed on the smartphone screen.

[0275] Step 5:

[0276] A user uploads financial report documents using a smartphone. The uploaded documents are sent to the server as text data.

[0277] Step 6:

[0278] The server analyzes the received financial report documents using text mining techniques (e.g., scikit-learn). Specifically, it extracts important information from the documents. The input is the text of the financial report documents, and the output is the extracted important information.

[0279] Step 7:

[0280] The server generates a summary based on the extracted key information. A generative AI model is used to generate a summary of the material. The input is the extracted key information, and the output is the generated summary text.

[0281] Step 8:

[0282] The server sends the generated summary to the smartphone, which displays the received summary to the user. The input is the generated summary text, and the output is the summary displayed on the smartphone screen.

[0283] In this way, users can enter questions and upload financial report documents via their smartphones and receive quick and accurate answers and summaries from the server.

[0284] Example 3

[0285] Next, a third embodiment of the third embodiment will be described. 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."

[0286] In the conventional system, it was difficult to answer questions from investors about indicators that were not included in financial reports, and the process of identifying appropriate sources, acquiring data, and generating answers using generative AI models was not automated, making it impossible to provide quick and accurate answers. This resulted in a decrease in investor satisfaction with their questions.

[0287] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0288] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering indicators other than financial reports, a means for answering questions from investors, a means for analyzing questions using natural language processing technology and identifying appropriate information sources, a means for acquiring data from information sources and generating answers using a generative AI model, and a means for providing the generated answers to investors, thereby making it possible to provide quick and accurate answers to questions from investors.

[0289] "Means of automatically responding to questions from investors" refers to a function whereby the system automatically generates and provides appropriate answers to questions entered by investors.

[0290] The "means of presenting summaries of financial reports and short messages, as well as points of concern" is a function that extracts important information from financial reports and short messages, and presents summaries and points of interest to investors.

[0291] "Means of answering questions about indicators other than financial statements" refers to a function that obtains information about indicators not listed in financial statements and provides answers to investors' questions.

[0292] "Means of answering investor questions" refers to the function whereby the system provides appropriate answers to questions that investors have.

[0293] "Means for analyzing questions using natural language processing technology and identifying appropriate information sources" is a function for analyzing questions from investors using natural language processing technology and identifying appropriate information sources for those questions.

[0294] "Means of obtaining data from an information source and generating an answer using a generative AI model" refers to a function that obtains necessary data from an identified information source and generates an answer using a generative AI model based on that data.

[0295] "Means of providing generated answers to investors" refers to a function that provides answers generated by a generative AI model to investors.

[0296] This invention relates to a system that automatically provides answers to questions from investors. The system operates in cooperation with a server, a terminal, and a user.

[0297] The server generates a program using the generative AI model. This program has the function of providing answers to questions from investors, even for indicators not included in financial reports. Specifically, the server analyzes questions using natural language processing technology, identifies appropriate sources, obtains data from the sources, and generates answers using the generative AI model.

[0298] In terms of hardware, the server is a computer system equipped with a high-performance processor and large-capacity memory. In terms of software, libraries such as spaCy and NLTK are used for natural language processing technology, and OpenAI's GPT-4 is used for generative AI models. MySQL (registered trademark) and PostgreSQL are used as database management systems (DBMS).

[0299] The terminal provides an interface for the user to input a question. The terminal communicates with the server through a web browser or a dedicated application, sends the question, and receives and displays the answer from the server.

[0300] A user (investor) inputs a question through the terminal interface. For example, "What is the level of customer satisfaction?" The terminal sends this question to the server. The server analyzes the question, identifies appropriate sources, retrieves data from the sources, and generates an answer using a generative AI model. The generated answer is provided to the user through the terminal.

[0301] As a concrete example, consider a case where an investor asks, "What is the level of customer satisfaction?" When this question is sent from the terminal to the server, the server retrieves the results of the customer satisfaction survey and uses a generative AI model to generate an answer like the following:

[0302] Example prompt:

[0303] Investors are asking: "What is the customer satisfaction rate?"

[0304] Server action:

[0305] 1. Analyze the questions to identify information about customer satisfaction.

[0306] 2. Obtain the results of the customer satisfaction survey via DBMS or API.

[0307] 3. Use a generative AI model (e.g. GPT-4) to generate an answer based on the acquired data.

[0308] 4. Provide the generated answers to the investors via the terminal.

[0309] Example of a generated answer:

[0310] "According to our most recent customer satisfaction survey, our customer satisfaction rate is 85%. The survey was conducted in March 2023 and covered 1,000 customers."

[0311] In this way, a system is realized in which the server, the terminals, and the users work together to provide prompt and accurate answers to questions from investors. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0312] Step 1:

[0313] The user enters a question.

[0314] A user (investor) inputs a question through the terminal interface. For example, the user inputs "What is the level of customer satisfaction?" The input question is temporarily stored in the terminal memory.

[0315] Step 2:

[0316] The terminal sends a question to the server.

[0317] The terminal sends the entered question to the server as an HTTP POST request. The request contains the question in JSON format. The input is the user's question, and the output is the request sent to the server.

[0318] Step 3:

[0319] The server parses the question.

[0320] The server analyzes the received HTTP request and extracts the question content. It uses a natural language processing (NLP) library (e.g., spaCy or NLTK) to analyze the intent of the question and identify the keyword "customer satisfaction." The input is the question data from the terminal, and the output is the analyzed keyword.

[0321] Step 4:

[0322] The server identifies the appropriate source of information.

[0323] The server identifies a database or external API for customer satisfaction surveys based on the analysis results. For example, it selects a table in a MySQL database where the results of the customer satisfaction surveys are stored. The input is the analyzed keywords, and the output is the identified information source.

[0324] Step 5:

[0325] The server retrieves the data from the information source.

[0326] The server retrieves data from the identified sources. For example, it might run a SQL query against a MySQL database to retrieve customer satisfaction survey results, or, in the case of an external API, it might send an HTTP GET request to an API endpoint to retrieve the data. The input is the identified sources and the output is the retrieved data.

[0327] Step 6:

[0328] The server generates an answer using a generative AI model.

[0329] The server inputs the acquired data into a generative AI model (e.g., GPT-4) to generate an answer in natural language. The generative AI model generates an answer based on the input data, taking into account appropriate context. The input is the acquired data, and the output is the generated answer.

[0330] Step 7:

[0331] The server sends the generated answer to the terminal.

[0332] The server sends the generated answer to the terminal as an HTTP response. The response includes the generated answer in JSON format. The input is the generated answer, and the output is the response sent to the terminal.

[0333] Step 8:

[0334] The terminal displays the answer to the user.

[0335] The terminal analyzes the received HTTP response and displays the response content on the user interface. The user can check the response through the terminal screen. The input is the response data from the server, and the output is the response display to the user.

[0336] (Application example 3)

[0337] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0338] With conventional investor information systems, it was difficult to provide appropriate answers to questions about KPIs that were not included in financial reports. In addition, there was a lack of a way to quickly and accurately provide detailed information that investors wanted, so investors' questions could not be resolved.

[0339] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means. In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and locations of points of concern, a means for answering KPIs other than the financial report, a means for answering questions from investors, a means for generating answers to questions from investors using a generative AI model, a means for acquiring data from an external information source, a means for generating a prompt sentence based on the acquired data, and a means for providing an answer based on the generated prompt sentence. This makes it possible to quickly and accurately provide detailed information desired by investors.

[0340] "Means of automatically responding to questions from investors" refers to a function whereby the system automatically generates and provides appropriate answers to questions entered by investors.

[0341] The "means of presenting summaries of financial reports and short messages as well as points of concern" is a function that extracts important information from financial reports and short messages, and presents summaries and points of particular interest to investors.

[0342] "Means of answering KPIs other than financial statements" is a function that obtains information about KPIs that are not listed in financial statements and provides answers to investor questions.

[0343] "Means of answering investor questions" refers to the function whereby the system provides appropriate answers to questions that investors have.

[0344] "Means for generating answers to questions from investors using a generative AI model" refers to a function that uses a generative AI model to automatically generate answers to questions from investors.

[0345] "Means for acquiring data from external information sources" refers to the function by which the system acquires necessary information from external data sources.

[0346] "Means for generating prompt sentences based on acquired data" refers to a function that generates prompt sentences to be input into a generative AI model based on acquired data.

[0347] "Means for providing an answer based on the generated prompt sentence" refers to a function in which a generative AI model uses the generated prompt sentence to provide an appropriate answer.

[0348] The system for implementing the present invention includes a number of means for automatically providing answers to questions from investors. A specific embodiment of the system will be described below.

[0349] First, the server receives questions from investors. Investors input their questions using devices such as smartphones or computers. Once the questions are sent to the server, the server uses natural language processing technology to analyze the questions and generate appropriate answers.

[0350] Next, the server uses text mining technology to present summaries of financial reports and short reports, as well as points of interest, allowing it to extract information that is important to investors and present it as a summary.

[0351] In addition, the server retrieves data from external sources to obtain information on KPIs other than financial reports, such as customer satisfaction survey results and market trends. Based on the retrieved data, it generates prompt sentences to input into the generative AI model.

[0352] Based on the generated prompt, the generative AI model generates an answer to the investor's question. For example, if an investor asks, "What is the level of customer satisfaction?", the server retrieves the results of a customer satisfaction survey and generates a prompt like the following:

[0353] “Question: What is the customer satisfaction rate?

[0354] Context: Customer Satisfaction Survey Results Report Content

[0355] Based on this prompt, the generative AI model generates an appropriate answer and provides it to the investor.

[0356] The implementation of this system uses the following hardware and software:

[0357] Hardware: Servers, smartphones, computers

[0358] Software: Python, requests library, transformers library

[0359] The server uses the requests library to retrieve data from external APIs and the transformers library to run generative AI models, allowing it to quickly and accurately provide investors with the detailed information they need.

[0360] The flow of the specific process in the application example 3 will be described with reference to FIG.

[0361] Step 1:

[0362] A user inputs a question using a terminal. The input question is sent to a server. The input data is the user's question text. The output data is the question text sent to the server.

[0363] Step 2:

[0364] The server analyzes the received question text using natural language processing technology. Specifically, it understands the intent of the question and extracts keywords to generate an appropriate answer. The input data is the user's question text. The output data is the analyzed keywords and the intent of the question.

[0365] Step 3:

[0366] The server uses text mining technology to extract important information from financial reports and short messages. This allows the server to present important information to investors as summaries. The input data is the text data of financial reports and short messages. The output data is the extracted important information and its summary.

[0367] Step 4:

[0368] The server obtains data from external information sources to obtain information on KPIs other than financial statements. Specifically, it obtains data such as customer satisfaction survey results and market trends using APIs. The input data is the API endpoint of the external information source. The output data is the obtained KPI data.

[0369] Step 5:

[0370] Based on the acquired KPI data, the server generates a prompt sentence to be input to the generative AI model. Specifically, the server creates a prompt sentence by combining the question text and the acquired KPI data. The input data is the user's question text and the acquired KPI data. The output data is the generated prompt sentence.

[0371] Step 6:

[0372] The server uses the generated prompt sentence to allow the generative AI model to generate an answer to the investor's question. Specifically, the server inputs the prompt sentence into the generative AI model and obtains the answer generated by the model. The input data is the generated prompt sentence. The output data is the answer text generated by the generative AI model.

[0373] Step 7:

[0374] The server sends the generated answer text to the user's device. The user uses the device to check the answer sent from the server. The input data is the answer text generated by the generative AI model. The output data is the answer text displayed on the user's device.

[0375] Furthermore, an emotion engine that estimates the emotion of the user may be combined. That is, the identification processing unit 290 may estimate the emotion of the user using the emotion identification model 59, and perform identification processing using the emotion of the user.

[0376] "Example 1"

[0377] One embodiment of the present invention is a system for automatically answering questions from investors using an emotion engine. This system receives questions from investors and recognizes the emotions of the investors using an emotion engine. The emotion engine, for example, estimates the emotions from the context and wording of the investor's questions. Then, it generates an appropriate answer depending on the emotions. For example, if the system senses that the investor is feeling anxious, it generates an answer that gives a sense of security.

[0378] "Example 2"

[0379] Another embodiment of the present invention is a system for presenting summaries of financial report materials and short messages and points of concern using an emotion engine. This system acquires financial report materials and short messages and extracts important information using text mining technology. It then uses an emotion engine to recognize the emotions of investors and presents information according to those emotions. For example, if an investor feels excited, the system presents information that will maintain that excitement.

[0380] "Example 3"

[0381] Furthermore, a system for answering KPIs other than financial reports using an emotion engine is also an embodiment of the present invention. This system receives questions from investors and obtains necessary information from sources other than financial reports. It then uses the emotion engine to recognize the emotions of the investors and answers KPIs according to those emotions. For example, if an investor feels confused, the system provides a clear and concise answer that clears up the confusion.

[0382] The process flow of each embodiment will be described below.

[0383] "Example 1"

[0384] Step 1: The system receives a query from an investor.

[0385] Step 2: The sentiment engine infers sentiment from the context and wording of the investor's question.

[0386] Step 3: The system generates an appropriate response based on the results of the sentiment engine. For example, if the system senses that the investor is anxious, it generates a reassuring response.

[0387] "Example 2"

[0388] Step 1: The system retrieves financial report materials and short reports.

[0389] Step 2: The system uses text mining techniques to extract important information.

[0390] Step 3: The sentiment engine recognises investor sentiment.

[0391] Step 4: The system presents information based on the results of the sentiment engine. For example, if an investor feels excited, the system presents information that maintains that excitement.

[0392] "Example 3"

[0393] Step 1: The system receives a query from an investor.

[0394] Step 2: The system obtains the required information from sources other than financial statements.

[0395] Step 3: The sentiment engine recognises investor sentiment.

[0396] Step 4: The system will provide a KPI answer based on the result of the sentiment engine. For example, if the investor feels confused, the system will provide a clear and concise answer that will clear up the confusion.

[0397] Example 1

[0398] Next, a description will be given of Example 1 of the embodiment 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".

[0399] Conventional information provision systems for investors have difficulty in providing prompt and appropriate answers to questions from investors. In addition, they are unable to generate answers that take into account the emotions of investors, and are unable to fully alleviate investors' concerns and doubts. Furthermore, there is a lack of means to efficiently extract financial report materials and other important information and present them to investors. To solve these issues, there is a need for a system that can automatically provide appropriate answers to investor questions and generate answers that take into account the emotions of investors.

[0400] 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.

[0401] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and points of concern, a means for answering indicators other than financial reports, a means for recognizing investors' emotions and generating answers according to those emotions, a means for analyzing questions using natural language processing technology, and a means for generating appropriate answers using a generative AI model. This makes it possible to provide prompt and appropriate answers to questions from investors and generate answers that take into account the emotions of investors. In addition, financial report materials and other important information can be efficiently extracted and presented to investors.

[0402] An "investor" is an individual or legal entity that invests funds in the financial markets with the aim of earning profits.

[0403] A "question" is a sentence or phrase that an investor enters into the system to request information.

[0404] An "automated response mechanism" is a feature that allows the system to generate responses to investor questions without human intervention.

[0405] A "financial report" is a document prepared by a company at regular intervals to report on its financial status and performance.

[0406] A "summary" is a concise summary of important information extracted from the financial report materials.

[0407] "Areas of concern" are those parts of the financial report that contain information that deserves particular attention.

[0408] An "indicator" is a number or data used to evaluate a company's performance and financial situation.

[0409] "Emotion recognition" is a function that infers emotions from the context and wording of investors' questions.

[0410] "Natural language processing technology" is a technology that enables computers to understand and analyze human language.

[0411] A "generative AI model" is an algorithm that uses artificial intelligence to generate appropriate answers based on input data.

[0412] "Means of analysis" refers to a function that uses natural language processing technology to analyze questions from investors and identify their intent.

[0413] "Means for generating an answer" is a function for generating an appropriate answer based on the analysis results.

[0414] This invention relates to a chatbot system for investors, which automatically provides appropriate answers to questions from investors and generates answers that take into account the emotions of investors. This system operates in cooperation with a server, a terminal, and a user.

[0415] Hardware and Software Used

[0416] Hardware: High-performance servers (e.g. virtual servers from a cloud service provider)

[0417] software:

[0418] Natural Language Processing: Natural Language Processing API

[0419] Emotion Recognition: Emotion Recognition Engine

[0420] Answer Generation: Generative AI Model

[0421] Specific operation of the system

[0422] 1. A user uses a terminal (PC or smartphone) to input a question to the chatbot. For example, the user might input, "What are the main points of the latest financial report?"

[0423] 2. The device sends this question to the server in real time.

[0424] 3. The server stores the received question in a database and sends it to a natural language processing API for analysis.

[0425] 4. The natural language processing API analyzes the question's grammatical structure and keywords and returns the analysis results to the server.

[0426] 5. The server identifies the intent of the question based on the analysis results and sends it to a generative AI model to generate an appropriate answer.

[0427] 6. The generative AI model generates an appropriate answer based on the analysis results and returns it to the server, for example, "The main points of the latest financial report are sales growth, improved profit margins, and new product launches."

[0428] 7. The server stores the generated answer in a database and sends it to the terminal.

[0429] 8. The terminal displays the received answer to the user, who then confirms the displayed answer.

[0430] Specific operation of emotion recognition

[0431] 1. A user types into the chatbot interface, "I'm worried about recent market movements."

[0432] 2. The device sends this question to the server in real time.

[0433] 3. The server stores the received questions in a database and sends them to an emotion recognition engine to recognize emotions.

[0434] 4. The emotion recognition engine infers the emotion from the question's context and wording, and returns the emotion recognition results to the server.

[0435] 5. Based on the emotion recognition results, the server determines that "the investor is feeling anxious" and sends this to a generative AI model to generate an answer corresponding to the emotion.

[0436] 6. The generative AI model generates an answer such as, "There is no need to worry about recent market movements. In the long term, your investments are stable," and returns it to the server.

[0437] 7. The server stores the generated answer in a database and sends it to the terminal.

[0438] 8. The terminal displays the received answer to the user, who then checks the displayed answer and feels reassured.

[0439] Examples of prompt statements

[0440] "What are the key takeaways from the latest earnings report?"

[0441] "I'm worried about recent market movements."

[0442] In this way, it is possible to provide prompt and appropriate answers to questions from investors and generate answers that take into account the sentiment of investors.

[0443] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0444] Step 1:

[0445] A user uses a terminal to type a question into the chatbot, for example, "What are the key takeaways from the latest earnings report?"

[0446] Input: The question entered by the user

[0447] Output: Question data sent by the device to the server

[0448] Step 2:

[0449] The terminal transmits the entered question to the server in real time.

[0450] Input: Question data entered by the user

[0451] Output: The question data sent to the server

[0452] Step 3:

[0453] The server stores the received questions in a database.

[0454] Input: Question data sent from the terminal

[0455] Output: Question data stored in a database

[0456] Step 4:

[0457] The server sends the question data to a natural language processing API for analysis.

[0458] Input: Question data stored in the database

[0459] Output: The parsing request sent to the natural language processing API

[0460] Step 5:

[0461] The natural language processing API analyzes the question's grammatical structure and keywords and returns the analysis results to the server.

[0462] Input: The analysis request sent by the server

[0463] Output: Analysis of the question's grammatical structure and keywords

[0464] Step 6:

[0465] The server uses the analysis results to identify the intent of the question and sends it to a generative AI model to generate an appropriate answer.

[0466] Input: Analysis results returned from the natural language processing API

[0467] Output: The prompt sent to the generative AI model

[0468] Step 7:

[0469] The generative AI model generates an appropriate answer based on the analysis results and returns it to the server, for example, "The main points of the latest financial report are sales growth, improved profit margins, and new product launches."

[0470] Input: Prompt text sent from the server

[0471] Output: The generated answer

[0472] Step 8:

[0473] The server stores the generated answer in a database and transmits it to the terminal.

[0474] Input: The answer returned by the generative AI model

[0475] Output: Answer data stored in the database, answer data sent to the device

[0476] Step 9:

[0477] The terminal displays the received answer to the user, who then checks the displayed answer.

[0478] Input: Answer data sent from the server

[0479] Output: The answer that is displayed to the user

[0480] Specific processing steps for emotion recognition

[0481] Step 1:

[0482] A user types into the chatbot's interface: "I'm worried about recent market movements."

[0483] Input: The question entered by the user

[0484] Output: Question data sent by the device to the server

[0485] Step 2:

[0486] The terminal transmits this question to the server in real time.

[0487] Input: Question data entered by the user

[0488] Output: The question data sent to the server

[0489] Step 3:

[0490] The server stores the received questions in a database.

[0491] Input: Question data sent from the terminal

[0492] Output: Question data stored in a database

[0493] Step 4:

[0494] The server transmits the question data to an emotion recognition engine to recognize emotions.

[0495] Input: Question data stored in the database

[0496] Output: Emotion recognition request sent to the emotion recognition engine

[0497] Step 5:

[0498] The emotion recognition engine infers emotions from the context and wording of the question and returns the emotion recognition results to the server.

[0499] Input: Emotion recognition request sent by the server

[0500] Output: Emotion recognition result

[0501] Step 6:

[0502] Based on the emotion recognition results, the server determines that "the investor is feeling anxious" and sends this to a generative AI model to generate an answer that corresponds to the emotion.

[0503] Input: Emotion recognition result returned by the emotion recognition engine

[0504] Output: The prompt sent to the generative AI model

[0505] Step 7:

[0506] The generative AI model generates an answer such as, "There is no need to worry about recent market movements. In the long term, your investments are stable," and returns it to the server.

[0507] Input: Prompt text sent from the server

[0508] Output: The generated answer

[0509] Step 8:

[0510] The server stores the generated answer in a database and transmits it to the terminal.

[0511] Input: The answer returned by the generative AI model

[0512] Output: Answer data stored in the database, answer data sent to the device

[0513] Step 9:

[0514] The terminal displays the received answer to the user, who then checks the displayed answer and feels reassured.

[0515] Input: Answer data sent from the server

[0516] Output: The answer that is displayed to the user

[0517] (Application example 1)

[0518] Next, a description will be given of Application Example 1 of Form 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."

[0519] Conventional chatbot systems for investors can provide appropriate answers to questions from investors, but they have the problem of not being able to fully resolve investors' concerns and doubts because they cannot generate answers that take into account investors' emotions. In addition, when extracting and presenting important information from financial report materials and short messages, there is also the problem of a lack of information provided that meets investors' interests.

[0520] 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.

[0521] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering key performance indicators other than financial reports, a means for analyzing emotions, and a means for adjusting answers based on emotions. This makes it possible to provide appropriate answers that take into account the emotions of investors and resolve their concerns and questions.

[0522] "Means of automatically responding to questions from investors" refers to a function that analyzes questions entered by investors and automatically generates and provides appropriate answers.

[0523] The "means of presenting summaries of financial reports and short messages, as well as points of interest" is a function that extracts important information from financial reports and short messages, and presents summaries of those documents and points of interest to investors.

[0524] "Means of responding to key performance indicators other than financial statements" refers to a function that provides information on indicators that are not included in financial statements but are important for evaluating a company's performance.

[0525] "Means of answering investor questions" is a function that provides appropriate answers to questions that investors have.

[0526] "Sentiment analysis means" is a function that analyzes the emotions expressed by investors' questions and comments and determines whether they are positive or negative.

[0527] The "means for adjusting responses based on emotions" is a function for generating appropriate responses according to the emotions of investors based on the emotional information obtained by the emotion analysis means.

[0528] A system for implementing this invention includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short notes and points of concern, a means for answering key performance indicators other than the financial report, a means for emotion analysis, and a means for adjusting answers based on emotions.

[0529] Hardware and Software Used

[0530] Hardware: Smartphone

[0531] Software: Python, OpenAI API, TextBlob

[0532] Data processing and calculation

[0533] An automated way to answer investor questions

[0534] The server receives questions entered by investors via their smartphones. The questions are analyzed using natural language processing technology and appropriate answers are generated. In this case, answers to questions are generated using the OpenAI API.

[0535] A means of presenting the summary of financial report documents and short reports, as well as points of concern

[0536] The server uses text mining technology to analyze financial reports and short messages and extract important information, which is then presented to investors as a summary.

[0537] Means of responding to key performance indicators other than financial statements

[0538] The server provides information on indicators that are not included in the financial statements but are important for evaluating the company's performance. This information is presented appropriately in response to investors' questions.

[0539] Emotion analysis means

[0540] The server uses TextBlob to analyze the sentiment of the investor's questions and comments, and based on the analysis, determines whether the question is positive or negative.

[0541] A way to tailor your answers based on emotions

[0542] The server generates an appropriate response according to the emotion of the investor based on the emotion information obtained by the emotion analysis means. For example, if the investor is feeling anxious, the server generates a response that gives the investor a sense of security.

[0543] Examples

[0544] Q: "What are some recent market trends?"

[0545] Answer: "The market has been very volatile recently, with technology stocks being particularly hard hit."

[0546] Examples of prompt statements

[0547] Answer investor questions: What have been the recent market trends?

[0548] In this way, the server can provide appropriate answers to investors' questions in real time and respond according to the investors' emotions.

[0549] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0550] Step 1:

[0551] A user enters an investment question using a smartphone.

[0552] Input: The question text entered by the user.

[0553] Output: The question text sent to the server.

[0554] Step 2:

[0555] The server analyzes the received question text using natural language processing technology.

[0556] Input: The question text submitted by the user

[0557] Output: The parsed meaning or intent of the question

[0558] Step 3:

[0559] Based on the analysis results, the server generates an appropriate answer using the OpenAI API.

[0560] Input: The parsed meaning or intent of the question.

[0561] Output: The generated answer text

[0562] Step 4:

[0563] The server uses TextBlob to parse sentiment from the user's question text.

[0564] Input: The question text submitted by the user

[0565] Output: Analyzed sentiment information (positive, negative, etc.)

[0566] Step 5:

[0567] The server then tailors the generated answer based on the sentiment analysis results, for example generating a reassuring answer if the user is feeling anxious.

[0568] Input: Generated answer text, parsed sentiment information

[0569] Output: Answer text adjusted according to sentiment

[0570] Step 6:

[0571] The server analyzes financial report documents and short messages using text mining technology to extract important information.

[0572] Input: Text data of financial reports and short messages

[0573] Output: Summary of key information extracted

[0574] Step 7:

[0575] A server provides information regarding key performance indicators other than financial statements.

[0576] Input: The user's question text

[0577] Output: Key Performance Indicator Answer Text

[0578] Step 8:

[0579] The server sends the generated answer text to the user's smartphone.

[0580] Input: Adjusted answer text

[0581] Output: Answer text displayed on the user's smartphone

[0582] In this way, through the specific actions performed at each step, it is possible to provide appropriate answers to the user's questions and respond in accordance with the user's emotions.

[0583] Example 2

[0584] Next, a second embodiment of the second embodiment will be described. 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."

[0585] Conventional investor information systems have difficulty quickly and accurately extracting important information from financial report materials and short messages and presenting it to investors in an appropriate format. In addition, they do not provide information based on investors' emotions, so they are unable to fully meet investors' needs. Furthermore, they lack the functionality to automatically generate appropriate answers to investors' questions.

[0586] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering indicators other than financial reports, a means for answering questions from investors, a means for recognizing investors' emotions using an emotion engine and presenting information according to the emotions, and a means for extracting important information from financial report materials and short messages using text mining technology. This makes it possible to provide investors with information quickly and accurately, and also makes it possible to present information according to investors' emotions. In addition, appropriate answers can be automatically generated to questions from investors.

[0587] An "investor" is an individual or entity that provides funds to a company or financial product in the hope of receiving a profit in return.

[0588] "Means for automatically answering questions" refers to technologies or systems that analyze questions from investors and generate appropriate answers.

[0589] A "financial report" is a document prepared by a company at regular intervals to report on its financial status and business performance.

[0590] A "tanshin" is a report that briefly summarizes a company's performance and financial status.

[0591] A "summary" is a concise summary that extracts important information from a long piece of text or report.

[0592] "Means of presenting sections of information of particular interest to investors" refers to technologies or systems for extracting and presenting information of particular interest to investors from financial report documents and short messages.

[0593] An "indicator" is a number or data used to evaluate a company's performance and financial situation.

[0594] An "emotion engine" is a technology or system that analyzes a user's emotions and provides information according to those emotions.

[0595] "Text mining technology" is a technology for extracting useful information from large amounts of text data.

[0596] "Means for presenting information" refers to technologies and systems for displaying extracted information to users in an easy-to-understand manner.

[0597] This invention relates to an information provision system for investors, and in particular to a system that extracts important information from financial report materials and short messages and presents it to investors in an appropriate format. Furthermore, it also has a function of providing information according to the sentiments of investors.

[0598] System Configuration

[0599] This system mainly consists of three elements: a server, a terminal, and a user.

[0600] server

[0601] The server has the following functions:

[0602] 1. Acquisition of financial report materials and short messages: The server acquires financial report materials and short messages from the official websites of companies and financial information providers. Specifically, it sends an HTTP request to download PDF files and saves them in local storage.

[0603] 2. Information extraction through text mining: The server converts the acquired materials into text format and extracts important information using natural language processing libraries "NLTK" and "SpaCy."

[0604] 3. Information summary generation: The server generates a summary based on the extracted information. The summary includes important indicators such as sales and profit margins. Specifically, the server uses the Gensim library to generate the summary text.

[0605] 4. Sentiment analysis by sentiment engine: The server uses IBM Watson(R) Tone Analyzer and Microsoft(R) Azure(R) Text Analytics to analyze investor sentiment. It obtains investor feedback and comments and sends them to the sentiment engine.

[0606] 5. Information display based on emotions: The server displays appropriate information to investors based on the results of emotion analysis. For example, if an investor is excited, it will select positive information and display it through the web interface or mobile app.

[0607] Terminal

[0608] The terminal has the role of displaying the information received from the server to the user, specifically through a web browser or a mobile app.

[0609] User

[0610] The users are investors who use the information provided by the system. The users can ask the system questions, and the server automatically generates answers to those questions.

[0611] Examples

[0612] Example 1: Extract sales and profit margins from financial report documents and present summary information such as "Sales for the first quarter of fiscal year 2023 were 100 million yen, and profit margins were 15%."

[0613] Example 2: If the sentiment engine detects that investors are excited, it will present positive information such as "New product launch is expected to significantly increase sales."

[0614] Examples of prompt statements

[0615] "Extract key indicators from financial report documents and generate summary information."

[0616] "Analyze sentiment using a sentiment engine to present information according to investor sentiment."

[0617] In this way, the server is a system that uses text mining technology and an emotion engine to extract important information from financial report documents and short messages, and presents appropriate information to investors.

[0618] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0619] Step 1:

[0620] Acquisition of financial report materials and short reports

[0621] Input: The URL of the company's official website or financial information service.

[0622] Specific behavior: The server sends an HTTP request to download the PDF file from the specified URL. For example, the server retrieves the PDF file from the URL "https: / / example.com / financial_report.pdf".

[0623] Data processing: Save the downloaded PDF file to local storage.

[0624] Output: A PDF file saved in local storage.

[0625] Step 2:

[0626] Information Extraction through Text Mining

[0627] Input: A PDF file saved in local storage.

[0628] What it does: The server converts PDF files to text format using the PyPDF2 library, then analyzes the text using NLTK and SpaCy to extract key metrics like sales and profit margins.

[0629] Data processing: The data converted to text format is analyzed using natural language processing libraries to extract important information.

[0630] Output: Extracted key information (e.g. sales, profit margins).

[0631] Step 3:

[0632] Generate a summary of information

[0633] Input: Extracted key information.

[0634] Specific operation: The server uses the "Gensim" library to summarize the extracted information and generate a summary sentence, such as "Sales in the first quarter of fiscal year 2023 were 100 million yen, and the profit margin was 15%."

[0635] Data processing: Organize the extracted information and generate a summary.

[0636] Output: The generated summary.

[0637] Step 4:

[0638] Sentiment analysis with emotion engine

[0639] Input: Investor feedback and comments.

[0640] How it works: The server sends investor feedback and comments to IBM Watson Tone Analyzer and Microsoft Azure Text Analytics for sentiment analysis.

[0641] Data processing: Send feedback and comments to the sentiment engine and receive sentiment analysis results.

[0642] Output: Emotion analysis results (e.g. excitement, relief, anxiety).

[0643] Step 5:

[0644] Information presentation based on emotions

[0645] Input: Sentiment analysis results and generated summary sentences.

[0646] How it works: Based on the results of sentiment analysis, the server selects appropriate information for investors and displays it through a web interface or mobile app. For example, if an investor is excited, it will present positive information such as "Sales are expected to increase significantly due to the launch of a new product."

[0647] Data processing: Select information based on the results of sentiment analysis and summaries, and convert it into a display format.

[0648] Output: Information presented to investors.

[0649] In this way, the server processes and analyzes the input data at each step, and ultimately provides appropriate information to investors.

[0650] (Application example 2)

[0651] Next, a description will be given of Application Example 2 of Form 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."

[0652] In conventional investment support systems, it was difficult for investors to quickly extract and understand important information from financial reports and short messages. In addition, information was not presented according to investors' emotions, and appropriate information was not provided according to investors' needs. Furthermore, there was a lack of real-time information notifications, so investors were unable to receive important information in a timely manner.

[0653] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0654] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering indicators other than financial reports, a means for answering questions from investors, a means for recognizing investors' emotions and presenting information according to those emotions, and a means for notifying investors in real time. This enables investors to quickly understand important information and receive appropriate information according to their emotions. Also, by notifying information in real time, investors can receive important information in a timely manner.

[0655] "Means of automatically responding to questions from investors" refers to a function that analyzes questions entered by investors, generates appropriate answers, and automatically provides them.

[0656] "Means of presenting summaries of financial reports and short messages and points of concern" is a function that extracts important information from financial reports and short messages, summarizes it, and presents it to investors.

[0657] "Means of answering questions about indicators other than financial statements" is a function that provides answers to questions about important indicators that are not included in financial statements.

[0658] "Means of answering investor questions" is a function that provides appropriate answers to questions that investors have.

[0659] "Means of recognizing investor emotions and presenting information in accordance with those emotions" refers to a function that recognizes investor emotions using an emotion engine and provides information in accordance with those emotions.

[0660] "Means of notifying investors in real time" refers to a function that notifies investors of important information in real time.

[0661] The system for carrying out this invention is composed of a server and an investor's terminal. The server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering indicators other than financial reports, a means for answering questions from investors, a means for recognizing investor emotions and presenting information according to those emotions, and a means for notifying investors in real time.

[0662] Hardware and Software Used

[0663] Hardware: Servers, smartphones

[0664] Software: Python, requests library, textblob library, transformers library

[0665] Data processing and calculation

[0666] 1. Automated ways to answer investor questions:

[0667] The server uses natural language processing technology to analyze questions from investors and generate appropriate answers. Specifically, it uses a Python natural language processing library to analyze questions, retrieves relevant information from a database, and generates answers.

[0668] 2. How to present the summary of financial report documents and brief notes and points of concern:

[0669] The server uses text mining techniques to extract important information from financial reports and short reports, summarizes them, and presents them to investors. Specifically, it uses the transformers library to summarize the text.

[0670] 3. For non-financial report metrics, please provide:

[0671] The server provides answers to questions about important metrics that aren't included in financial reports, including pulling data from external databases and APIs.

[0672] 4. How investors can get answers to their questions:

[0673] The server provides appropriate answers to questions that investors have by analyzing the questions using natural language processing technology and providing relevant information.

[0674] 5. Ways to recognise investor sentiment and present information according to that sentiment:

[0675] The server uses the textblob library to recognize the sentiment of the investor and provide information according to that sentiment, for example, if the investor is excited, it provides positive information.

[0676] 6. Means of real-time investor notification:

[0677] The server notifies investors of important information in real time, including via push notifications and email notifications.

[0678] Examples

[0679] For example, if an investor types, "I'm very excited about the recent financial results!", the server will use the textblob library to recognize positive sentiment and provide positive information. It will also generate a summary of the earnings call and notify the investor in real time.

[0680] Examples of prompt statements

[0681] "Generate a summary of financial statements."

[0682] "Recognize investor sentiment and provide information that responds to that sentiment."

[0683] In this way, a system can be realized that helps investors obtain information more effectively.

[0684] The flow of the identification process in the application example 2 will be described with reference to FIG.

[0685] Step 1:

[0686] A user inputs a question using a smartphone. The input question is sent to the server. The input data is text in a natural language format.

[0687] Step 2:

[0688] The server analyzes user questions using natural language processing technology. Specifically, it uses a Python natural language processing library to tokenize the questions and perform semantic analysis. The input data is the user's question text, and the output data is the semantic information of the analyzed question.

[0689] Step 3:

[0690] The server retrieves relevant information from the database based on the semantic information of the parsed question. Specifically, it uses an SQL query to extract the required information from the database. The input data is the semantic information of the parsed question, and the output data is the information retrieved from the database.

[0691] Step 4:

[0692] The server generates an appropriate answer based on the acquired information. Specifically, it uses a generative AI model to convert the acquired information into an answer in natural language format. The input data is the information acquired from the database, and the output data is the generated answer text.

[0693] Step 5:

[0694] The server transmits the generated answer to the user's smartphone. The user can check the answer on the smartphone. The input data is the generated answer text, and the output data is the answer displayed on the user's smartphone.

[0695] Step 6:

[0696] The server retrieves financial reports and short messages and extracts important information using text mining techniques. Specifically, it summarizes the text using the transformers library. The input data is the text of financial reports and short messages, and the output data is the summarized important information.

[0697] Step 7:

[0698] The server uses the textblob library to recognize the user's emotion. Specifically, the server analyzes the user's input text and calculates an emotion score. The input data is the user's input text, and the output data is the emotion score.

[0699] Step 8:

[0700] The server presents emotion-specific information based on the recognized emotion score. For example, if a positive emotion is recognized, it provides positive information. The input data are the emotion score and the summarized important information, and the output data is emotion-specific information.

[0701] Step 9:

[0702] The server notifies the user of important information in real time. Specifically, the server sends the information to the user using push notifications or email notifications. The input data is the information according to the emotion, and the output data is the notification sent to the user.

[0703] Example 3

[0704] Next, a third embodiment of the third embodiment will be described. 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."

[0705] Conventional investor information systems have difficulty in providing appropriate answers to questions about key performance indicators (KPIs) that are not included in financial statements. In addition, they provide uniform answers without considering investors' emotions, which means that investors' doubts and concerns cannot be fully resolved. Furthermore, question analysis using natural language processing technology and emotion analysis using emotion recognition technology are insufficient, which can result in a decline in the quality of answers provided to investors.

[0706] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0707] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering key performance indicators other than the financial report, a means for analyzing investor emotions using emotion recognition technology and generating answers according to the emotions, and a means for analyzing questions from investors using natural language processing technology and generating appropriate answers. This makes it possible to provide appropriate answers to questions from investors using information other than the financial report and generate answers according to investor emotions.

[0708] An "investor" is an individual or legal entity that provides capital to a company or project and expects to receive a profit in return.

[0709] "Means of automatically answering questions" refers to a function in which the system receives questions from investors and automatically generates and provides appropriate answers.

[0710] An "financial report" is a formal document that reports a company's financial condition and performance, usually issued on a quarterly or annual basis.

[0711] A "tanshin" is a report that briefly summarizes a company's financial results, and is intended to provide quick information to investors and other interested parties.

[0712] An "abstract" is a concise summary of the main points of a longer document or report.

[0713] "Means of presenting sections of concern" is a function that extracts and presents important information and sections of interest to investors from financial report documents and short messages.

[0714] A "key performance indicator (KPI)" is a specific indicator used to evaluate a company's business performance, and includes sales, profit margins, customer satisfaction, etc.

[0715] "Emotion recognition technology" is a technology for analyzing and recognizing emotions from a user's input and actions.

[0716] "Natural language processing technology" is a technology that enables computers to understand and analyze human language, and is used to analyze the intent of questions.

[0717] "Analysis" is the process of examining data or information in detail to understand its structure and meaning.

[0718] The "means for generating an appropriate answer" is a function that automatically generates the most appropriate answer to the user's question based on the analysis results.

[0719] This invention is a system that automatically provides answers to questions from investors, and includes technology for generating summaries of financial report materials and short messages, answers to key performance indicators (KPIs) other than financial reports, and answers according to investor sentiment. Specific embodiments of this system are described below.

[0720] Hardware and Software Used

[0721] Servers: Database servers, Web servers

[0722] Software: Natural Language Processing (NLP) engine, emotion recognition engine, database management system (DBMS)

[0723] System Configuration

[0724] 1. The user (investor) uses a terminal to input a question.

[0725] 2. The terminal sends the entered question to the server.

[0726] 3. The server receives the question and analyzes it using an NLP engine.

[0727] 4. Based on the analysis results, the server obtains necessary data from sources other than financial reports (for example, the results of customer satisfaction surveys).

[0728] 5. The server uses an emotion recognition engine to analyze the user's emotions.

[0729] 6. Based on the emotion analysis results, the server generates an appropriate answer according to the user's emotions.

[0730] 7. The server sends the generated response to the terminal.

[0731] 8. The terminal displays the answer to the user.

[0732] Examples

[0733] For example, explain what happens if an investor asks, "What is the customer satisfaction rate?"

[0734] 1. A user types into a terminal, "What is your customer satisfaction rating?"

[0735] 2. The device sends this question to the server.

[0736] 3. The server uses an NLP engine to analyze the question and understands that you are seeking information about "customer satisfaction."

[0737] 4. The server retrieves the customer satisfaction survey results from the database.

[0738] 5. The server uses an emotion recognition engine to analyze whether the user is confused or not.

[0739] 6. If the server determines that the user is confused, it generates a clear, concise answer to clear up the confusion.

[0740] 7. The server generates an answer such as "According to the latest customer satisfaction survey, customer satisfaction is 85%" and sends it to the terminal.

[0741] 8. The terminal displays this answer to the user.

[0742] Examples of prompt statements

[0743] Below is an example of a prompt sentence to be input to the generative AI model.

[0744] Investors ask: "What is the customer satisfaction rate?"

[0745] The system responds: "According to our most recent customer satisfaction survey, customer satisfaction is 85%."

[0746] In this way, the system provides appropriate answers to the questions of the investors. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0747] Step 1:

[0748] The user inputs a question into the terminal.

[0749] Input: A user types "What is your customer satisfaction rating?" into an input field.

[0750] Output: The entered question is stored in the device's memory.

[0751] Specific operation: The user inputs a question using the device's keyboard or touch screen. The device displays the input in real time and the user confirms the question by pressing the Enter key or clicking the Send button.

[0752] Step 2:

[0753] The terminal sends a question to the server.

[0754] Input: The question entered by the user.

[0755] Output: The question is sent to the server as an HTTP request.

[0756] Specific operation: The terminal converts the entered question into JSON format and sends it as an HTTP POST request to the server's API endpoint.

[0757] Step 3:

[0758] The server receives the question and analyzes it using an NLP engine.

[0759] Input: A question sent from the device (e.g., "What is your customer satisfaction level?").

[0760] Output: Analysis of the question intent (e.g., asking for information about "customer satisfaction").

[0761] Specific operation: The server parses the received JSON data and uses an NLP engine (e.g., SpaCy or BERT) to analyze the intent of the question. As a result of the analysis, it extracts the keywords and intent of the question.

[0762] Step 4:

[0763] The server retrieves the required data.

[0764] Input: Analysis results from an NLP engine (e.g. "Customer Satisfaction").

[0765] Output: The results of a customer satisfaction survey retrieved from a database (e.g., "Customer satisfaction is 85%").

[0766] Specific operation: Based on the analysis results, the server queries the database management system (DBMS) for related data. For example, it retrieves the latest survey results on "customer satisfaction" from the database.

[0767] Step 5:

[0768] The server analyzes the user's emotions using an emotion recognition engine.

[0769] Input: A user question (e.g., "What is the customer satisfaction rate?").

[0770] Output: User sentiment analysis (e.g. "I'm confused").

[0771] Specific operation: The server uses an emotion recognition engine (for example, IBM Watson or Microsoft Azure's emotion analysis API) to analyze the emotion from the user's question. As a result of the analysis, it determines whether the user is confused.

[0772] Step 6:

[0773] The server generates a response according to the emotion.

[0774] Input: Sentiment analysis results (e.g. "I'm confused") and information obtained from a database (e.g. "Customer satisfaction is 85%").

[0775] Output: An appropriate answer depending on the user's sentiment (e.g. "According to our latest customer satisfaction survey, customer satisfaction is 85%").

[0776] Specific operation: The server generates an answer according to the user's emotions based on the results of emotion analysis. For example, if the server determines that the user is confused, it generates a simple and clear answer.

[0777] Step 7:

[0778] The server sends the response to the terminal.

[0779] Input: The generated answer (e.g. "According to our most recent customer satisfaction survey, customer satisfaction is 85%.").

[0780] Output: The answer is sent to the device as an HTTP response.

[0781] Specific operation: The server converts the generated answer into JSON format and sends it to the terminal as an HTTP response.

[0782] Step 8:

[0783] The terminal displays the answer to the user.

[0784] Input: The answer sent by the server (e.g. "According to our most recent customer satisfaction survey, customer satisfaction is 85%.").

[0785] Output: The answer that is displayed to the user.

[0786] Specific operation: The terminal analyzes the received JSON data and displays it on the screen in a format that is easy for the user to see. For example, it displays "According to the latest customer satisfaction survey, customer satisfaction is 85%."

[0787] (Application example 3)

[0788] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0789] Conventional investor information systems have difficulty in providing appropriate answers to questions about KPIs that are not included in financial statements. In addition, they provide uniform answers without considering investors' feelings, which means that investors' doubts and concerns cannot be fully resolved. Furthermore, they are not convenient because they do not take into account the use on mobile devices such as smartphones.

[0790] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means. In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering KPIs other than the financial report, a means for recognizing investor emotions using an emotion engine and providing an answer according to the emotion, and an application installed on a smartphone. This makes it possible to provide appropriate KPI information from information sources other than the financial report in response to investor questions and provide an answer according to the investor's emotions. In addition, since it can be used on a smartphone, convenience is improved.

[0791] An "investor" is an individual or legal entity that provides funds to a company or project with the aim of receiving a profit in return.

[0792] A "question" is an inquiry that an investor poses to the system to clarify the information or concerns they have.

[0793] "Automatic response means" refers to a function in which the system automatically generates and provides appropriate responses to questions from investors.

[0794] A "financial report" is an official document that reports a company's financial status and performance.

[0795] A "tanshin" is a document that provides a concise report of a company's quarterly performance.

[0796] A "summary" is information that concisely summarizes the main points of financial reports or short reports.

[0797] "Parts of concern" are those parts that contain information that investors are particularly interested in.

[0798] "Presentation means" refers to the functionality by which the system displays information to investors.

[0799] "KPI" stands for Key Performance Indicator and is an index used to measure a company's performance and degree of goal achievement.

[0800] An "emotion engine" is a technology that recognizes investor emotions and responds accordingly.

[0801] A "smartphone" is a mobile phone that can connect to the Internet and use applications.

[0802] An "application" is software installed on a smartphone to provide specific functions or services.

[0803] To implement this invention, the following system configuration is necessary. The system includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering KPIs other than financial reports, a means for recognizing investor emotions using an emotion engine and answering according to those emotions, and an application installed on a smartphone.

[0804] Hardware and Software Configuration

[0805] Hardware:

[0806] Smartphone

[0807] software:

[0808] Python

[0809] Transformers Library (Hugging Face)

[0810] Requests Library

[0811] Data processing and calculation flow

[0812] server:

[0813] 1. Receive questions from investors.

[0814] 2. The question content is analyzed using natural language processing technology.

[0815] 3. Obtain KPI information relevant to your question from sources other than financial statements.

[0816] 4. Recognize investor sentiment using a sentiment engine.

[0817] 5. Generate answers that respond to investor sentiment.

[0818] Device (smartphone):

[0819] 1. Investors enter their questions through the application.

[0820] 2. Receive the response from the server and display it to the investor.

[0821] Examples

[0822] For example, if an investor asks, "What is the customer satisfaction rate?", the server retrieves the results of a customer satisfaction survey and uses an emotion engine to recognize the investor's emotions. If the investor is determined to be confused, the server generates a clear and concise answer such as, "To resolve your confusion, we provide the following information: Customer satisfaction rate is 85%," and displays it on the smartphone.

[0823] Examples of prompt statements

[0824] Q: What is the customer satisfaction rate?

[0825] Emotion: Confusion

[0826] Answer: To clear up any confusion, I provide the following information: Our customer satisfaction rate is 85%.

[0827] In this way, it is possible to provide appropriate KPI information to investors' questions and respond according to their emotions. In addition, the system can be used on smartphones, improving convenience.

[0828] The flow of the specific process in the application example 3 will be described with reference to FIG.

[0829] Step 1:

[0830] A user starts the application on their smartphone and inputs a question. The input question is sent as text data from the device to the server.

[0831] Step 2:

[0832] The server analyzes the received question using natural language processing technology. Specifically, it understands the intent of the question and extracts related keywords. The results of this analysis are used in the next step.

[0833] Step 3:

[0834] Based on the analysis results, the server retrieves relevant KPI information from sources other than financial reports. For example, if the question is about customer satisfaction, it retrieves the results of a customer satisfaction survey. This information can be retrieved from a database or an external API.

[0835] Step 4:

[0836] The server recognizes the user's emotion using an emotion engine based on the acquired KPI information. The emotion engine receives the question text data as input and outputs the user's emotion (e.g., confusion, relief, excitement, etc.).

[0837] Step 5:

[0838] The server generates an appropriate answer depending on the perceived emotion. For example, if the user is perceived as confused, the server generates a clear and concise answer such as "To clear up your confusion, we provide the following information: Customer satisfaction is 85%."

[0839] Step 6:

[0840] The server sends the generated answer to the terminal. The terminal displays the received answer to the user. The user can check the answer on the screen of their smartphone.

[0841] In this way, it is possible to provide appropriate KPI information to investors' questions and respond emotionally.

[0842] 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 a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice 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 voice data.

[0843] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by making a neural network perform deep learning. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image is input. The data generation model 58 performs inference on the input inference data according to 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.

[0844] Another example of generative AI is Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0845] In the above embodiment, an example was given in which the specific process was performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0846] [Second embodiment]

[0847] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0848] 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.

[0849] 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 wide area network (WAN) and / or a local area network (LAN).

[0850] 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.

[0851] 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 the voice according to instructions from the processor 46.

[0852] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a 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 (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0853] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54. The transmission and reception of various types of information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0854] Fig. 4 shows an example of 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.

[0855] 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.

[0856] 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.

[0857] 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.

[0858] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0859] "Example 1"

[0860] One embodiment of the present invention is a chatbot system for investors. The system receives questions from investors and analyzes the content of the questions using natural language processing technology. Based on the analysis results, the system generates an appropriate answer and provides it to the investor. For example, if an investor asks, "What are the main points of the latest financial report?", the system analyzes the financial report materials, extracts the main points, and answers the investor.

[0861] "Example 2"

[0862] Another embodiment of the present invention is a function for presenting summaries of financial report materials and short messages and points of concern. This function uses text mining technology to extract important information from financial report materials and short messages and present it to investors. For example, important indicators such as sales and profit margins listed in financial report materials are extracted and summarized information is provided to investors.

[0863] "Example 3"

[0864] Furthermore, another embodiment of the present invention has a function for answering KPIs other than financial reports. This function provides answers to questions from investors about KPIs that are not listed in financial reports. For example, if an investor asks, "What is the level of customer satisfaction?", the system obtains information from sources other than financial reports, such as the results of a customer satisfaction survey, and answers the investor.

[0865] The process flow of each embodiment will be described below.

[0866] "Example 1"

[0867] Step 1: The system receives the investor's question.

[0868] Step 2: The system uses natural language processing techniques to analyze the content of the question.

[0869] Step 3: Based on the analysis results, the system generates an appropriate answer.

[0870] Step 4: Provide the generated responses to investors.

[0871] "Example 2"

[0872] Step 1: The system retrieves financial report materials and short reports.

[0873] Step 2: Use text mining techniques to extract important information from the acquired materials.

[0874] Step 3: Summarize the extracted information and present it to investors.

[0875] "Example 3"

[0876] Step 1: The system receives the investor's question.

[0877] Step 2: The system obtains the required information from sources other than financial statements.

[0878] Step 3: Based on the information obtained, the system answers the investor's questions.

[0879] Example 1

[0880] Next, a description will be given of Example 1 of Form 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".

[0881] Conventional investor relations systems had the problem that it was difficult to quickly and accurately respond to questions from investors. In addition, there was a lack of a way to efficiently extract information on financial reports and other indicators and provide it to investors. This meant that investors had to spend a lot of time and effort to obtain the information they needed.

[0882] 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.

[0883] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting a summary of financial report materials and important points, a means for answering indicators other than financial reports, a means for analyzing questions using natural language processing technology and generating appropriate answers, a means for generating answers using a generative AI model, a means for generating prompt sentences and inputting them into the generative AI model, and a means for providing the generated answers to investors, thereby enabling investors to quickly and accurately obtain necessary information.

[0884] "Investor" refers to an individual or entity that invests money in stocks, bonds, and other financial instruments in the financial markets.

[0885] "Means of automatically answering questions" refers to the function of receiving questions from investors and having the system automatically generate and provide appropriate answers.

[0886] "Financial report materials" refers to reports on financial status and performance that are issued periodically by a company.

[0887] "Means of presenting summaries and key points" refers to the function of extracting important information from financial report documents, summarizing it concisely and presenting it to investors.

[0888] An "indicator" refers to numerical data or statistical information used to evaluate a company's performance and financial situation.

[0889] "Natural language processing technology" refers to technology that enables computers to understand and analyze human language.

[0890] A "generative AI model" refers to a model for generating text data using artificial intelligence.

[0891] A "prompt sentence" refers to an instruction sentence to be input into a generative AI model.

[0892] "Analysis" refers to the process of breaking down input data and understanding its meaning and structure.

[0893] "Means for generating an answer" refers to the function of creating an appropriate answer based on the analysis results.

[0894] "Means of providing" refers to the functionality for communicating the generated answers to investors.

[0895] This invention relates to a chatbot system for investors, which provides quick and accurate answers to questions from investors. This system operates among three parties: a server, a terminal, and a user.

[0896] The server receives questions from investors and analyzes them using natural language processing technology, for example using a generative AI model such as OpenAI's GPT-4. The server extracts the intent and keywords of the question and obtains information to generate an appropriate answer.

[0897] The server generates an appropriate answer based on the analysis results. For example, the server refers to the latest financial report documents stored in the database and extracts the main points. In this case, the server inputs the following prompt sentences to the generative AI model:

[0898] Prompt: "What are the key takeaways from your latest earnings report?"

[0899] The generative AI model will generate an answer based on this prompt, for example in the following format:

[0900] Answer: "The main takeaways from our most recent financial report were that revenue increased 10% year over year and net income increased 15%. We also saw strong sales of our new products, which gives us good prospects for future growth."

[0901] The server transmits the generated answer to the terminal. The terminal receives the answer transmitted from the server and displays it to the user. The user can check the answer on the screen of the terminal.

[0902] The specific hardware configuration of this system is that the server is a computer equipped with a high-performance processor and large-capacity memory, and the terminal is a device operated by the user, such as a PC or smartphone. The software includes a natural language processing engine, generative AI model, and database management system that run on the server side.

[0903] As a concrete example, consider the case where a user inputs a question such as "What are the main points of the latest financial report?" The device sends this question to the server, which analyzes it using GPT-4. The server retrieves the latest financial report materials from the database and extracts the main points. The server inputs the prompt sentence into the generative AI model and sends the generated answer to the device. The device displays the answer to the user, allowing the user to quickly obtain the information they need.

[0904] Thus, the present invention provides a system that allows investors to quickly and accurately obtain the information they need.

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

[0906] Step 1:

[0907] The user enters a question.

[0908] A user inputs a question about an investment into a terminal (PC or smartphone). For example, the user inputs, "What are the main points of the latest financial report?" The input question is stored as text data on the terminal.

[0909] Step 2:

[0910] The terminal sends a question to the server.

[0911] The terminal transmits the question input by the user to the server. At this time, the terminal transmits text data to the server via the Internet. The input data is the question text of the user.

[0912] Step 3:

[0913] The server receives the query.

[0914] The server receives the question sent from the terminal. The received question is stored in the server as text data. The input data is the user's question text sent from the terminal.

[0915] Step 4:

[0916] The server parses the question.

[0917] The server analyzes the received question using natural language processing technology. For this analysis, it uses a generative AI model such as OpenAI's GPT-4. The server extracts the intent and keywords of the question and obtains information to generate an appropriate answer. The input data is the user's question text, and the output data is the analysis result.

[0918] Step 5:

[0919] The server generates an appropriate response.

[0920] The server generates an appropriate answer based on the analysis results. For example, the server refers to the latest financial report documents stored in the database and extracts the main points. In this case, the server inputs the following prompt sentences to the generative AI model:

[0921] Prompt: "What are the key takeaways from your latest earnings report?"

[0922] The generative AI model generates an answer based on this prompt. The input data is the prompt and the analysis result, and the output data is the generated answer.

[0923] Step 6:

[0924] The server sends the response to the terminal.

[0925] The server transmits the generated answer to the terminal. At this time, the server transmits the answer to the terminal as text data. The input data is the generated answer text, and the output data is the answer text transmitted to the terminal.

[0926] Step 7:

[0927] The terminal displays the answer to the user.

[0928] The terminal receives the answer sent from the server and displays it to the user. The user can check the answer on the screen of the terminal. The input data is the answer text sent from the server, and the output data is the answer text displayed to the user.

[0929] (Application example 1)

[0930] Next, a description will be given of Application Example 1 of Form 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".

[0931] Conventional chatbot systems for investors have the ability to automatically answer questions from investors, but they lack the ability to provide appropriate answers to security-related questions.In addition, they have issues with not improving the user experience because they do not provide a user interface or retrieve information from a database efficiently.

[0932] 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.

[0933] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering performance indicators other than financial reports, a means for answering investors' concerns, a means for generating appropriate answers to security-related questions, a means for analyzing questions and generating appropriate answers using natural language processing technology, a means for acquiring information from a database and providing it to users, and a means for providing a user interface. This makes it possible to provide quick and accurate answers to questions from investors and provide comprehensive support including security-related information.

[0934] "Means of automatically responding to questions from investors" refers to a function that analyzes questions entered by investors and automatically generates and provides appropriate answers.

[0935] The "means of presenting summaries of financial report documents and short messages as well as points of interest" is a function that extracts important information from financial report documents and short messages and presents the user with summaries and points of particular interest.

[0936] The "means of answering questions about performance indicators other than financial statements" is a function that provides appropriate answers to questions about performance indicators other than financial statements.

[0937] "Means of answering investor questions" is a function that provides appropriate answers to questions that investors have.

[0938] The "means for generating appropriate answers to security-related questions" is a function for analyzing security questions and generating and providing appropriate answers.

[0939] "Means of analyzing questions using natural language processing technology and generating appropriate answers" refers to a function that uses natural language processing technology to analyze a user's question and generate an appropriate answer based on the content of the question.

[0940] "Means for retrieving information from a database and providing it to a user" refers to a function for searching information stored in a database, retrieving the necessary information, and providing it to a user.

[0941] "Means for providing a user interface" refers to a function that provides an interface for a user to input a question and receive a response.

[0942] A system for implementing this invention includes means for automatically answering questions from investors, means for presenting summaries of financial report materials and short messages and points of concern, means for answering performance indicators other than financial reports, means for answering investor concerns, means for generating appropriate answers to security-related questions, means for analyzing questions and generating appropriate answers using natural language processing technology, means for retrieving information from a database and providing it to a user, and means for providing a user interface.

[0943] System Configuration

[0944] The server includes a question analysis engine using natural language processing technology, a database access module, an answer generation engine, and a user interface module. These modules are implemented using Python and natural language processing libraries (e.g., spaCy and Transformers).

[0945] Program processing explanation

[0946] When the server receives a question from a user, it first analyzes the question using natural language processing technology. Natural language processing libraries such as spaCy and Transformers are used for this analysis. The analyzed question is used to retrieve related information from a database via a database access module. The database contains financial report documents, performance indicators, security-related information, and so on.

[0947] The acquired information is processed into appropriate answers by the answer generation engine, which uses a generative AI model to generate answers, and the generated answers are provided to the user through a user interface module.

[0948] Examples

[0949] For example, if a user asks, "What are the latest security threats?" the system would do the following:

[0950] 1. The user interface module receives a question.

[0951] 2. Analyze the question using natural language processing techniques.

[0952] 3. The database access module retrieves the relevant security information from the database.

[0953] 4. The answer generation engine generates appropriate answers based on the information obtained.

[0954] 5. The user interface module provides the generated answers to the user.

[0955] Examples of prompt statements

[0956] User: What are the latest security threats?

[0957] Systems: The latest security threat is ransomware attacks.

[0958] In this way, the system can provide quick and accurate answers to questions from investors and users, providing comprehensive support, including security-related information.

[0959] The flow of the specific process in the application example 1 will be described with reference to FIG.

[0960] Step 1:

[0961] The user inputs a question. The user inputs a question through the user interface of a smartphone or a robot. The input question is sent to the server.

[0962] Step 2:

[0963] The server receives the question. The server receives the question sent by the user and analyzes the question using natural language processing technology. Specifically, it uses natural language processing libraries such as spaCy and Transformers to analyze the meaning of the question. The input is the user's question, and the output is the structure of the analyzed question.

[0964] Step 3:

[0965] The server retrieves information from the database. Based on the content of the parsed question, the server searches the database for relevant information. The database may contain financial report documents, performance indicators, security-related information, etc. The input is the parsed question structure, and the output is the relevant information.

[0966] Step 4:

[0967] The server generates an answer. The server uses a generative AI model to generate an appropriate answer based on the acquired information. Specifically, the acquired information is input as a prompt into the generative AI model to generate an appropriate answer. The input is the information acquired from the database, and the output is the generated answer.

[0968] Step 5:

[0969] The server provides the answer to the user. The server provides the generated answer to the user through a user interface module. The user can check the answer on the screen of a smartphone or a robot. The input is the generated answer, and the output is the answer provided to the user.

[0970] Specifically, when a user asks "What are the latest security threats?", the server receives the question, analyzes it using natural language processing technology, and retrieves information about the latest security threats from a database. It then uses a generative AI model to generate the answer "The latest security threat is a ransomware attack" and provides it to the user.

[0971] Example 2

[0972] Next, a description will be given of Example 2 of the embodiment 2. In the following description, the data processing device 12 is referred to as a "server" and the smart glasses 214 are referred to as a "terminal".

[0973] In the conventional system, investors had to manually extract and summarize important information from financial reports and short messages, which was cumbersome and time-consuming. It was also difficult to quickly and appropriately respond to questions from investors, so an efficient means of resolving investors' concerns was needed.

[0974] 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.

[0975] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering questions about indicators other than financial reports, a means for uploading financial report materials and short messages, a means for receiving uploaded materials, a means for analyzing materials using a text mining tool, a means for extracting important indicators, a means for summarizing the extracted information, a means for formatting the summary in natural language using a generative AI model, a means for transmitting the formatted summary information to a terminal, and a means for the terminal to present the summary information to a user. This makes it possible to quickly and accurately provide information required by investors and efficiently resolve investors' questions.

[0976] An "investor" is an individual or legal entity that provides capital to a company or project in the hope of receiving a profit in return.

[0977] "Means for automatically answering questions" refers to a part of a system that has the function of receiving questions from investors, analyzing them using natural language processing technology, and generating appropriate answers.

[0978] A "financial report" is a document prepared by a company at regular intervals to report on its financial status and business performance.

[0979] A "tanshin" is a report that briefly summarizes a company's financial results.

[0980] A summary is a short sentence that concisely summarizes a longer piece of text or complex information.

[0981] "Means of presenting sections of points of concern" refers to part of a system that has the function of extracting and presenting important information and sections of interest to investors from financial report documents and short messages.

[0982] An "indicator" is a number or data used to evaluate a company's performance and financial situation.

[0983] An "uploading means" is a part of the system that has the functionality to allow a user to send files to the system.

[0984] The "receiving means" is a part of the system in which the server has the function of receiving and storing files sent by users.

[0985] A "text mining tool" is software for analyzing text data and extracting important information.

[0986] "Means for analysis" refers to a part of a system that has the function of analyzing the contents of documents using text mining tools.

[0987] The "means for extracting important indicators" is a part of the system that has the function of extracting important data such as sales revenue and profit margins from the analysis results.

[0988] The "summarizing means" is the part of the system that has the functionality to generate a concise summary based on the extracted key information.

[0989] A "generative AI model" is a model for generating sentences in natural language using artificial intelligence technology.

[0990] A "means for formatting a summary in natural language" is a part of a system that has the function of formatting summary information in easy-to-understand natural language using a generative AI model.

[0991] A "terminal" is a device through which a user accesses the system and receives information.

[0992] A "means for presenting" is a part of the system where a terminal has the ability to display information to a user.

[0993] This invention is a system that allows investors to quickly and accurately extract important information from financial reports and short messages, and provides summary information. The system operates in cooperation with a server, a terminal, and a user.

[0994] Hardware and Software Used

[0995] Hardware: Servers, terminals

[0996] Software: Text mining tools (e.g. NLTK, SpaCy), generative AI models (e.g. GPT-4)

[0997] System Operation Overview

[0998] 1. The user uploads financial report documents and short messages to the system.

[0999] 2. The server receives the uploaded materials and analyzes the text data using text mining tools.

[1000] 3. The server extracts key indicators such as sales revenue and profit margins from the analysis results.

[1001] 4. The server summarizes the extracted information and uses a generative AI model to format it in natural language for investors to understand.

[1002] 5. The server sends the formatted summary information to the terminal and presents it to the user.

[1003] Examples

[1004] For example, if a user uploads "Financial report materials for Q1 2023" to the system, the server will process it as follows:

[1005] 1. The user uploads the "2023 Q1 Financial Report Materials" to the system.

[1006] 2. The server receives the material and analyzes the text data using NLTK.

[1007] 3. The server extracts key indicators such as sales revenue and profit margins.

[1008] 4. The server summarizes the extracted information and uses GPT-4 to generate a natural language summary such as, "Revenue in Q1 2023 increased 10% compared to the same period last year, and profit margins reached 15%."

[1009] 5. The server sends the generated summary information to the terminal and presents it to the user.

[1010] Examples of prompt statements

[1011] An example of a prompt to be input to the generative AI model is as follows:

[1012] "Please extract and summarize key indicators from the text below. Text: 'In the financial results report for Q1 2023, sales increased 10% year-on-year and profit margins reached 15%. In addition, sales of new products are strong, and future growth is expected.'"

[1013] Using this prompt, the generative AI model extracts key information and generates a summary.

[1014] This system allows investors to quickly obtain the information they need, enabling them to make efficient investment decisions. It also allows for quick and appropriate responses to investor questions, resolving investors' doubts and providing highly reliable information.

[1015] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1016] Step 1:

[1017] The user uploads financial report documents and short messages to the system.

[1018] Specific operation: The user selects "Financial report materials for Q1 2023.pdf" and clicks the upload button.

[1019] Input: The financial report file selected by the user.

[1020] Output: The uploaded file is sent to the server.

[1021] Step 2:

[1022] The server receives the uploaded material.

[1023] Specific operation: The server receives "2023Q1Financial Report Materials.pdf" and saves it in the / uploads directory.

[1024] Input: Financial report file submitted by user.

[1025] Output: The file saved in the specified directory on the server.

[1026] Step 3:

[1027] The server analyzes the material using text mining tools.

[1028] Specific operation: The server uses NLTK to extract text from "2023Q1Financial Report Materials.pdf" and begins analysis.

[1029] Input: Financial report file stored on the server.

[1030] Output: The extracted text data.

[1031] Step 4:

[1032] The server extracts key indicators.

[1033] Specific operation: The server extracts important indicators based on keywords such as "sales revenue," "profit margin," and "year-on-year comparison."

[1034] Input: The extracted text data.

[1035] Output: Extracted key indicators data.

[1036] Step 5:

[1037] The server summarizes the extracted information.

[1038] Specific operation: The server generates summary information based on the extracted important indicators.

[1039] Input: Extracted key indicators data.

[1040] Output: Summarized information.

[1041] Step 6:

[1042] The server uses a generative AI model to format the summary in natural language.

[1043] How it works: The server inputs a prompt sentence into GPT-4, which generates a natural language summary such as, "Sales in Q1 2023 increased 10% year-on-year, and profit margins reached 15%."

[1044] Input: The summarized information and the prompt.

[1045] Output: Natural language formatted summary information.

[1046] Step 7:

[1047] The server transmits the formatted summary information to the terminal.

[1048] Specific operation: The server sends the generated summary information to the user's terminal as an HTTP response.

[1049] Input: A formatted summary in natural language.

[1050] Output: Summary information sent to the user's terminal.

[1051] Step 8:

[1052] The terminal presents the summary information to the user.

[1053] Specific behavior: The device displays on the browser, "Revenue in the first quarter of 2023 increased 10% year-on-year, and profit margin reached 15%."

[1054] Input: Summary information sent by the server.

[1055] Output: Summary information displayed to the user.

[1056] (Application example 2)

[1057] Next, a description will be given of Application Example 2 of Form 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".

[1058] Conventional investor information systems have difficulty quickly and accurately extracting and summarizing important information from financial reports and short messages. They also lack the ability to automatically generate appropriate answers to questions from investors. This means that investors have to spend a lot of time and effort to obtain the information they need.

[1059] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means. In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering indicators other than the financial report, and a means for extracting important information from the financial report materials and generating summaries through an application installed on a smartphone. This enables investors to quickly and accurately obtain the information they need.

[1060] An "investor" is an individual or legal entity that provides funds to a company or project with the aim of receiving a profit in return.

[1061] "Means for automatically answering questions" refers to technological means for analyzing questions from investors and generating appropriate answers.

[1062] A "financial report" is an official document that reports a company's financial status and performance.

[1063] A "tanshin" is a document that provides a concise report of a company's quarterly performance and financial status.

[1064] A "summary" is a short summary of a longer document or information, extracting and concisely presenting the important points.

[1065] "Means of presenting the sections of information of particular interest to investors" refers to technological means for extracting and presenting information of particular interest to investors from financial report documents and short messages.

[1066] An "indicator" is a specific number or data used to evaluate a company's performance and financial situation.

[1067] A "smartphone" is a mobile phone that can connect to the Internet and install applications.

[1068] An "Application" is a software program designed to provide a particular function or service.

[1069] "Text mining technology" is a data analysis technique for extracting useful information from large amounts of text data.

[1070] "Natural language processing technology" is a technology that enables computers to understand and analyze human language.

[1071] "Material information" refers to key data and indicators contained in financial reports and short reports that investors need to make decisions.

[1072] A "means for generating a summary" is a technical means for extracting important information from a long document and summarizing it concisely.

[1073] A system for implementing the present invention includes a server, a smartphone, and related software components. A specific configuration and operation of the system will be described below.

[1074] System Configuration

[1075] 1. Server:

[1076] The server uses natural language processing and text mining technologies to analyze questions from investors and generate appropriate answers.

[1077] The server extracts important information from financial reports and short messages and generates summaries.

[1078] The server then sends the generated summary and answers to the smartphone.

[1079] 2. Smartphone:

[1080] The smartphone displays the summary and answers sent from the server through a dedicated application.

[1081] A user can use a smartphone to input a question and send it to the server.

[1082] 3. Software Components:

[1083] A natural language processing library (e.g. spaCy): Used to parse investor questions and generate appropriate answers.

[1084] Text mining libraries (e.g. scikit-learn): Used to extract key information from financial reports and short reports.

[1085] Data processing and calculation

[1086] 1. Question analysis and answer generation:

[1087] The server receives questions from investors and analyzes them using natural language processing techniques.

[1088] Based on the parsed question, an appropriate answer is generated.

[1089] The generated answers are sent to the smartphone and displayed to the user.

[1090] 2. Generate summary of financial report documents:

[1091] The server receives financial report documents and short messages and uses text mining technology to extract important information.

[1092] A summary is generated based on the extracted information.

[1093] The generated summary is sent to the smartphone and displayed to the user.

[1094] Examples

[1095] Text from the financial report: "Financial report for fiscal year 2023. Sales increased 10% year-on-year, and profit margins improved 5%. Sales of new products were strong, with growth in the Asian market being particularly notable."

[1096] Example prompt: "Please extract and summarise the key information from this financial statement."

[1097] This system allows investors to quickly and accurately obtain the information they need, improving the efficiency of their decision-making. In addition, by using a smartphone, investors can enjoy the convenience of being able to access information anytime, anywhere.

[1098] The flow of the specific process in the application example 2 will be described with reference to FIG.

[1099] Step 1:

[1100] A user inputs a question using a smartphone and sends it to the server. The input question is sent to the server as text data.

[1101] Step 2:

[1102] The server analyzes the received question using natural language processing technology (e.g., spaCy). Specifically, it analyzes the grammatical structure of the question and extracts important keywords and phrases. The input is the user's question text, and the output is the analyzed keywords and phrases.

[1103] Step 3:

[1104] The server generates appropriate answers based on the analyzed keywords and phrases. It uses a generative AI model to generate the best answer to the question. The input is the analyzed keywords and phrases, and the output is the generated answer text.

[1105] Step 4:

[1106] The server sends the generated answer to the smartphone, which displays the received answer to the user. The input is the generated answer text, and the output is the answer displayed on the smartphone screen.

[1107] Step 5:

[1108] A user uploads financial report documents using a smartphone. The uploaded documents are sent to the server as text data.

[1109] Step 6:

[1110] The server analyzes the received financial report documents using text mining techniques (e.g., scikit-learn). Specifically, it extracts important information from the documents. The input is the text of the financial report documents, and the output is the extracted important information.

[1111] Step 7:

[1112] The server generates a summary based on the extracted key information. A generative AI model is used to generate a summary of the material. The input is the extracted key information, and the output is the generated summary text.

[1113] Step 8:

[1114] The server sends the generated summary to the smartphone, which displays the received summary to the user. The input is the generated summary text, and the output is the summary displayed on the smartphone screen.

[1115] In this way, users can enter questions and upload financial report documents via their smartphones and receive quick and accurate answers and summaries from the server.

[1116] Example 3

[1117] Next, a third embodiment of the third embodiment will be described. 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".

[1118] In the conventional system, it was difficult to answer questions from investors about indicators that were not included in financial reports, and the process of identifying appropriate sources, acquiring data, and generating answers using generative AI models was not automated, making it impossible to provide quick and accurate answers. This resulted in a decrease in investor satisfaction with their questions.

[1119] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1120] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering indicators other than financial reports, a means for answering questions from investors, a means for analyzing questions using natural language processing technology and identifying appropriate information sources, a means for acquiring data from information sources and generating answers using a generative AI model, and a means for providing the generated answers to investors, thereby making it possible to provide quick and accurate answers to questions from investors.

[1121] "Means of automatically responding to questions from investors" refers to a function whereby the system automatically generates and provides appropriate answers to questions entered by investors.

[1122] The "means of presenting summaries of financial reports and short messages, as well as points of concern" is a function that extracts important information from financial reports and short messages, and presents summaries and points of interest to investors.

[1123] "Means of answering questions about indicators other than financial statements" refers to a function that obtains information about indicators not listed in financial statements and provides answers to investors' questions.

[1124] "Means of answering investor questions" refers to the function whereby the system provides appropriate answers to questions that investors have.

[1125] "Means for analyzing questions using natural language processing technology and identifying appropriate information sources" is a function for analyzing questions from investors using natural language processing technology and identifying appropriate information sources for those questions.

[1126] "Means of obtaining data from an information source and generating an answer using a generative AI model" refers to a function that obtains necessary data from an identified information source and generates an answer using a generative AI model based on that data.

[1127] "Means of providing generated answers to investors" refers to a function that provides answers generated by a generative AI model to investors.

[1128] This invention relates to a system that automatically provides answers to questions from investors. The system operates in cooperation with a server, a terminal, and a user.

[1129] The server generates a program using the generative AI model. This program has the function of providing answers to questions from investors, even for indicators not included in financial reports. Specifically, the server analyzes questions using natural language processing technology, identifies appropriate sources, obtains data from the sources, and generates answers using the generative AI model.

[1130] In terms of hardware, the server is a computer system equipped with a high-performance processor and large memory capacity. In terms of software, libraries such as spaCy and NLTK are used for natural language processing technology, and OpenAI's GPT-4 is used for generative AI models. MySQL, PostgreSQL, etc. are used as database management systems (DBMS).

[1131] The terminal provides an interface for the user to input a question. The terminal communicates with the server through a web browser or a dedicated application, sends the question, and receives and displays the answer from the server.

[1132] A user (investor) inputs a question through the terminal interface. For example, "What is the level of customer satisfaction?" The terminal sends this question to the server. The server analyzes the question, identifies appropriate sources, retrieves data from the sources, and generates an answer using a generative AI model. The generated answer is provided to the user through the terminal.

[1133] As a concrete example, consider a case where an investor asks, "What is the level of customer satisfaction?" When this question is sent from the terminal to the server, the server retrieves the results of the customer satisfaction survey and uses a generative AI model to generate an answer like the following:

[1134] Example prompt:

[1135] Investors are asking: "What is the customer satisfaction rate?"

[1136] Server action:

[1137] 1. Analyze the questions to identify information about customer satisfaction.

[1138] 2. Obtain the results of the customer satisfaction survey via DBMS or API.

[1139] 3. Use a generative AI model (e.g. GPT-4) to generate an answer based on the acquired data.

[1140] 4. Provide the generated answers to the investors via the terminal.

[1141] Example of a generated answer:

[1142] "According to our most recent customer satisfaction survey, our customer satisfaction rate is 85%. The survey was conducted in March 2023 and covered 1,000 customers."

[1143] In this way, a system is realized in which the server, the terminals, and the users work together to provide prompt and accurate answers to questions from investors. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1144] Step 1:

[1145] The user enters a question.

[1146] A user (investor) inputs a question through the terminal interface. For example, the user inputs "What is the level of customer satisfaction?" The input question is temporarily stored in the terminal memory.

[1147] Step 2:

[1148] The terminal sends a question to the server.

[1149] The terminal sends the entered question to the server as an HTTP POST request. The request contains the question in JSON format. The input is the user's question, and the output is the request sent to the server.

[1150] Step 3:

[1151] The server parses the question.

[1152] The server analyzes the received HTTP request and extracts the question content. It uses a natural language processing (NLP) library (e.g., spaCy or NLTK) to analyze the intent of the question and identify the keyword "customer satisfaction." The input is the question data from the terminal, and the output is the analyzed keyword.

[1153] Step 4:

[1154] The server identifies the appropriate source of information.

[1155] The server identifies a database or external API for customer satisfaction surveys based on the analysis results. For example, it selects a table in a MySQL database where the results of the customer satisfaction surveys are stored. The input is the analyzed keywords, and the output is the identified information source.

[1156] Step 5:

[1157] The server retrieves the data from the information source.

[1158] The server retrieves data from the identified sources. For example, it might run a SQL query against a MySQL database to retrieve customer satisfaction survey results, or, in the case of an external API, it might send an HTTP GET request to an API endpoint to retrieve the data. The input is the identified sources and the output is the retrieved data.

[1159] Step 6:

[1160] The server generates an answer using a generative AI model.

[1161] The server inputs the acquired data into a generative AI model (e.g., GPT-4) to generate an answer in natural language. The generative AI model generates an answer based on the input data, taking into account appropriate context. The input is the acquired data, and the output is the generated answer.

[1162] Step 7:

[1163] The server sends the generated answer to the terminal.

[1164] The server sends the generated answer to the terminal as an HTTP response. The response includes the generated answer in JSON format. The input is the generated answer, and the output is the response sent to the terminal.

[1165] Step 8:

[1166] The terminal displays the answer to the user.

[1167] The terminal analyzes the received HTTP response and displays the response content on the user interface. The user can check the response through the terminal screen. The input is the response data from the server, and the output is the response display to the user.

[1168] (Application example 3)

[1169] Next, a description will be given of Application Example 3 of Form Example 3. 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".

[1170] With conventional investor information systems, it was difficult to provide appropriate answers to questions about KPIs that were not included in financial reports. In addition, there was a lack of a way to quickly and accurately provide detailed information that investors wanted, so investors' questions could not be resolved.

[1171] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means. In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and locations of points of concern, a means for answering KPIs other than the financial report, a means for answering questions from investors, a means for generating answers to questions from investors using a generative AI model, a means for acquiring data from an external information source, a means for generating a prompt sentence based on the acquired data, and a means for providing an answer based on the generated prompt sentence. This makes it possible to quickly and accurately provide detailed information desired by investors.

[1172] "Means of automatically responding to questions from investors" refers to a function whereby the system automatically generates and provides appropriate answers to questions entered by investors.

[1173] The "means of presenting summaries of financial reports and short messages as well as points of concern" is a function that extracts important information from financial reports and short messages, and presents summaries and points of particular interest to investors.

[1174] "Means of answering KPIs other than financial statements" is a function that obtains information about KPIs that are not listed in financial statements and provides answers to investor questions.

[1175] "Means of answering investor questions" refers to the function whereby the system provides appropriate answers to questions that investors have.

[1176] "Means for generating answers to questions from investors using a generative AI model" refers to a function that uses a generative AI model to automatically generate answers to questions from investors.

[1177] "Means for acquiring data from external information sources" refers to the function by which the system acquires necessary information from external data sources.

[1178] "Means for generating prompt sentences based on acquired data" refers to a function that generates prompt sentences to be input into a generative AI model based on acquired data.

[1179] "Means for providing an answer based on the generated prompt sentence" refers to a function in which a generative AI model uses the generated prompt sentence to provide an appropriate answer.

[1180] The system for implementing the present invention includes a number of means for automatically providing answers to questions from investors. A specific embodiment of the system will be described below.

[1181] First, the server receives questions from investors. Investors input their questions using devices such as smartphones or computers. Once the questions are sent to the server, the server uses natural language processing technology to analyze the questions and generate appropriate answers.

[1182] Next, the server uses text mining technology to present summaries of financial reports and short reports, as well as points of interest, allowing it to extract information that is important to investors and present it as a summary.

[1183] In addition, the server retrieves data from external sources to obtain information on KPIs other than financial reports, such as customer satisfaction survey results and market trends. Based on the retrieved data, it generates prompt sentences to input into the generative AI model.

[1184] Based on the generated prompt, the generative AI model generates an answer to the investor's question. For example, if an investor asks, "What is the level of customer satisfaction?", the server retrieves the results of a customer satisfaction survey and generates a prompt like the following:

[1185] “Question: What is the customer satisfaction rate?

[1186] Context: Customer Satisfaction Survey Results Report Content

[1187] Based on this prompt, the generative AI model generates an appropriate answer and provides it to the investor.

[1188] The implementation of this system uses the following hardware and software:

[1189] Hardware: Servers, smartphones, computers

[1190] Software: Python, requests library, transformers library

[1191] The server uses the requests library to retrieve data from external APIs and the transformers library to run generative AI models, allowing it to quickly and accurately provide investors with the detailed information they need.

[1192] The flow of the specific process in the application example 3 will be described with reference to FIG.

[1193] Step 1:

[1194] A user inputs a question using a terminal. The input question is sent to a server. The input data is the user's question text. The output data is the question text sent to the server.

[1195] Step 2:

[1196] The server analyzes the received question text using natural language processing technology. Specifically, it understands the intent of the question and extracts keywords to generate an appropriate answer. The input data is the user's question text. The output data is the analyzed keywords and the intent of the question.

[1197] Step 3:

[1198] The server uses text mining technology to extract important information from financial reports and short messages. This allows the server to present important information to investors as summaries. The input data is the text data of financial reports and short messages. The output data is the extracted important information and its summary.

[1199] Step 4:

[1200] The server obtains data from external information sources to obtain information on KPIs other than financial statements. Specifically, it obtains data such as customer satisfaction survey results and market trends using APIs. The input data is the API endpoint of the external information source. The output data is the obtained KPI data.

[1201] Step 5:

[1202] Based on the acquired KPI data, the server generates a prompt sentence to be input to the generative AI model. Specifically, the server creates a prompt sentence by combining the question text and the acquired KPI data. The input data is the user's question text and the acquired KPI data. The output data is the generated prompt sentence.

[1203] Step 6:

[1204] The server uses the generated prompt sentence to allow the generative AI model to generate an answer to the investor's question. Specifically, the server inputs the prompt sentence into the generative AI model and obtains the answer generated by the model. The input data is the generated prompt sentence. The output data is the answer text generated by the generative AI model.

[1205] Step 7:

[1206] The server sends the generated answer text to the user's device. The user uses the device to check the answer sent from the server. The input data is the answer text generated by the generative AI model. The output data is the answer text displayed on the user's device.

[1207] In addition, 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.

[1208] "Example 1"

[1209] One embodiment of the present invention is a system for automatically answering questions from investors using an emotion engine. This system receives questions from investors and recognizes the emotions of the investors using an emotion engine. The emotion engine, for example, estimates the emotions from the context and wording of the investor's questions. Then, it generates an appropriate answer depending on the emotions. For example, if the system senses that the investor is feeling anxious, it generates an answer that gives a sense of security.

[1210] "Example 2"

[1211] Another embodiment of the present invention is a system for presenting summaries of financial report materials and short messages and points of concern using an emotion engine. This system acquires financial report materials and short messages and extracts important information using text mining technology. It then uses an emotion engine to recognize the emotions of investors and presents information according to those emotions. For example, if an investor feels excited, the system presents information that will maintain that excitement.

[1212] "Example 3"

[1213] Furthermore, a system for answering KPIs other than financial reports using an emotion engine is also an embodiment of the present invention. This system receives questions from investors and obtains necessary information from sources other than financial reports. It then uses the emotion engine to recognize the emotions of the investors and answers KPIs according to those emotions. For example, if an investor feels confused, the system provides a clear and concise answer that clears up the confusion.

[1214] The process flow of each embodiment will be described below.

[1215] "Example 1"

[1216] Step 1: The system receives a query from an investor.

[1217] Step 2: The sentiment engine infers sentiment from the context and wording of the investor's question.

[1218] Step 3: The system generates an appropriate response based on the results of the sentiment engine. For example, if the system senses that the investor is anxious, it generates a reassuring response.

[1219] "Example 2"

[1220] Step 1: The system retrieves financial report materials and short reports.

[1221] Step 2: The system uses text mining techniques to extract important information.

[1222] Step 3: The sentiment engine recognises investor sentiment.

[1223] Step 4: The system presents information based on the results of the sentiment engine. For example, if an investor feels excited, the system presents information that maintains that excitement.

[1224] "Example 3"

[1225] Step 1: The system receives a query from an investor.

[1226] Step 2: The system obtains the required information from sources other than financial statements.

[1227] Step 3: The sentiment engine recognises investor sentiment.

[1228] Step 4: The system will provide a KPI answer based on the result of the sentiment engine. For example, if the investor feels confused, the system will provide a clear and concise answer that will clear up the confusion.

[1229] Example 1

[1230] Next, a description will be given of Example 1 of Form 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".

[1231] Conventional information provision systems for investors have difficulty in providing prompt and appropriate answers to questions from investors. In addition, they are unable to generate answers that take into account the emotions of investors, and are unable to fully alleviate investors' concerns and doubts. Furthermore, there is a lack of means to efficiently extract financial report materials and other important information and present them to investors. To solve these issues, there is a need for a system that can automatically provide appropriate answers to investor questions and generate answers that take into account the emotions of investors.

[1232] 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.

[1233] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and points of concern, a means for answering indicators other than financial reports, a means for recognizing investors' emotions and generating answers according to those emotions, a means for analyzing questions using natural language processing technology, and a means for generating appropriate answers using a generative AI model. This makes it possible to provide prompt and appropriate answers to questions from investors and generate answers that take into account the emotions of investors. In addition, financial report materials and other important information can be efficiently extracted and presented to investors.

[1234] An "investor" is an individual or legal entity that invests funds in the financial markets with the aim of earning profits.

[1235] A "question" is a sentence or phrase that an investor enters into the system to request information.

[1236] An "automated response mechanism" is a feature that allows the system to generate responses to investor questions without human intervention.

[1237] A "financial report" is a document prepared by a company at regular intervals to report on its financial status and performance.

[1238] A "summary" is a concise summary of important information extracted from the financial report materials.

[1239] "Areas of concern" are those parts of the financial report that contain information that deserves particular attention.

[1240] An "indicator" is a number or data used to evaluate a company's performance and financial situation.

[1241] "Emotion recognition" is a function that infers emotions from the context and wording of investors' questions.

[1242] "Natural language processing technology" is a technology that enables computers to understand and analyze human language.

[1243] A "generative AI model" is an algorithm that uses artificial intelligence to generate appropriate answers based on input data.

[1244] "Means of analysis" refers to a function that uses natural language processing technology to analyze questions from investors and identify their intent.

[1245] "Means for generating an answer" is a function for generating an appropriate answer based on the analysis results.

[1246] This invention relates to a chatbot system for investors, which automatically provides appropriate answers to questions from investors and generates answers that take into account the emotions of investors. This system works in cooperation with a server, a terminal, and a user.

[1247] Hardware and Software Used

[1248] Hardware: High-performance servers (e.g. virtual servers from a cloud service provider)

[1249] software:

[1250] Natural Language Processing: Natural Language Processing API

[1251] Emotion Recognition: Emotion Recognition Engine

[1252] Answer Generation: Generative AI Model

[1253] Specific operation of the system

[1254] 1. A user uses a terminal (PC or smartphone) to input a question to the chatbot. For example, the user might input, "What are the main points of the latest financial report?"

[1255] 2. The device sends this question to the server in real time.

[1256] 3. The server stores the received question in a database and sends it to a natural language processing API for analysis.

[1257] 4. The natural language processing API analyzes the question's grammatical structure and keywords and returns the analysis results to the server.

[1258] 5. The server identifies the intent of the question based on the analysis results and sends it to a generative AI model to generate an appropriate answer.

[1259] 6. The generative AI model generates an appropriate answer based on the analysis results and returns it to the server, for example, "The main points of the latest financial report are sales growth, improved profit margins, and new product launches."

[1260] 7. The server stores the generated answer in a database and sends it to the terminal.

[1261] 8. The terminal displays the received answer to the user, who then confirms the displayed answer.

[1262] Specific operation of emotion recognition

[1263] 1. A user types into the chatbot interface, "I'm worried about recent market movements."

[1264] 2. The device sends this question to the server in real time.

[1265] 3. The server stores the received questions in a database and sends them to an emotion recognition engine to recognize emotions.

[1266] 4. The emotion recognition engine infers the emotion from the question's context and wording, and returns the emotion recognition results to the server.

[1267] 5. Based on the emotion recognition results, the server determines that "the investor is feeling anxious" and sends this to a generative AI model to generate an answer corresponding to the emotion.

[1268] 6. The generative AI model generates an answer such as, "There is no need to worry about recent market movements. In the long term, your investments are stable," and returns it to the server.

[1269] 7. The server stores the generated answer in a database and sends it to the terminal.

[1270] 8. The terminal displays the received answer to the user, who then checks the displayed answer and feels reassured.

[1271] Examples of prompt statements

[1272] "What are the key takeaways from the latest earnings report?"

[1273] "I'm worried about recent market movements."

[1274] In this way, it is possible to provide prompt and appropriate answers to questions from investors and generate answers that take into account the sentiment of investors.

[1275] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1276] Step 1:

[1277] A user uses a terminal to type a question into the chatbot, for example, "What are the key takeaways from the latest earnings report?"

[1278] Input: The question entered by the user

[1279] Output: Question data sent by the device to the server

[1280] Step 2:

[1281] The terminal transmits the entered question to the server in real time.

[1282] Input: Question data entered by the user

[1283] Output: The question data sent to the server

[1284] Step 3:

[1285] The server stores the received questions in a database.

[1286] Input: Question data sent from the terminal

[1287] Output: Question data stored in a database

[1288] Step 4:

[1289] The server sends the question data to a natural language processing API for analysis.

[1290] Input: Question data stored in the database

[1291] Output: The parsing request sent to the natural language processing API

[1292] Step 5:

[1293] The natural language processing API analyzes the question's grammatical structure and keywords and returns the analysis results to the server.

[1294] Input: The analysis request sent by the server

[1295] Output: Analysis of the question's grammatical structure and keywords

[1296] Step 6:

[1297] The server uses the analysis results to identify the intent of the question and sends it to a generative AI model to generate an appropriate answer.

[1298] Input: Analysis results returned from the natural language processing API

[1299] Output: The prompt sent to the generative AI model

[1300] Step 7:

[1301] The generative AI model generates an appropriate answer based on the analysis results and returns it to the server, for example, "The main points of the latest financial report are sales growth, improved profit margins, and new product launches."

[1302] Input: Prompt text sent from the server

[1303] Output: The generated answer

[1304] Step 8:

[1305] The server stores the generated answer in a database and transmits it to the terminal.

[1306] Input: The answer returned by the generative AI model

[1307] Output: Answer data stored in the database, answer data sent to the device

[1308] Step 9:

[1309] The terminal displays the received answer to the user, who then checks the displayed answer.

[1310] Input: Answer data sent from the server

[1311] Output: The answer that is displayed to the user

[1312] Specific processing steps for emotion recognition

[1313] Step 1:

[1314] A user types into the chatbot's interface: "I'm worried about recent market movements."

[1315] Input: The question entered by the user

[1316] Output: Question data sent by the device to the server

[1317] Step 2:

[1318] The terminal transmits this question to the server in real time.

[1319] Input: Question data entered by the user

[1320] Output: The question data sent to the server

[1321] Step 3:

[1322] The server stores the received questions in a database.

[1323] Input: Question data sent from the terminal

[1324] Output: Question data stored in a database

[1325] Step 4:

[1326] The server transmits the question data to an emotion recognition engine to recognize emotions.

[1327] Input: Question data stored in the database

[1328] Output: Emotion recognition request sent to the emotion recognition engine

[1329] Step 5:

[1330] The emotion recognition engine infers emotions from the context and wording of the question and returns the emotion recognition results to the server.

[1331] Input: Emotion recognition request sent by the server

[1332] Output: Emotion recognition result

[1333] Step 6:

[1334] Based on the emotion recognition results, the server determines that "the investor is feeling anxious" and sends this to a generative AI model to generate an answer that corresponds to the emotion.

[1335] Input: Emotion recognition result returned by the emotion recognition engine

[1336] Output: The prompt sent to the generative AI model

[1337] Step 7:

[1338] The generative AI model generates an answer such as, "There is no need to worry about recent market movements. In the long term, your investments are stable," and returns it to the server.

[1339] Input: Prompt text sent from the server

[1340] Output: The generated answer

[1341] Step 8:

[1342] The server stores the generated answer in a database and transmits it to the terminal.

[1343] Input: The answer returned by the generative AI model

[1344] Output: Answer data stored in the database, answer data sent to the device

[1345] Step 9:

[1346] The terminal displays the received answer to the user, who then checks the displayed answer and feels reassured.

[1347] Input: Answer data sent from the server

[1348] Output: The answer that is displayed to the user

[1349] (Application example 1)

[1350] Next, a description will be given of Application Example 1 of Form 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".

[1351] Conventional chatbot systems for investors can provide appropriate answers to questions from investors, but they have the problem of not being able to fully resolve investors' concerns and doubts because they cannot generate answers that take into account investors' emotions. In addition, when extracting and presenting important information from financial report materials and short messages, there is also the problem of a lack of information provided that meets investors' interests.

[1352] 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.

[1353] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering key performance indicators other than financial reports, a means for analyzing emotions, and a means for adjusting answers based on emotions. This makes it possible to provide appropriate answers that take into account the emotions of investors and resolve their concerns and questions.

[1354] "Means of automatically responding to questions from investors" refers to a function that analyzes questions entered by investors and automatically generates and provides appropriate answers.

[1355] The "means of presenting summaries of financial reports and short messages, as well as points of interest" is a function that extracts important information from financial reports and short messages, and presents summaries of those documents and points of interest to investors.

[1356] "Means of responding to key performance indicators other than financial statements" refers to a function that provides information on indicators that are not included in financial statements but are important for evaluating a company's performance.

[1357] "Means of answering investor questions" is a function that provides appropriate answers to questions that investors have.

[1358] "Sentiment analysis means" is a function that analyzes the emotions expressed by investors' questions and comments and determines whether they are positive or negative.

[1359] The "means for adjusting responses based on emotions" is a function for generating appropriate responses according to the emotions of investors based on the emotional information obtained by the emotion analysis means.

[1360] A system for implementing this invention includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short notes and points of concern, a means for answering key performance indicators other than the financial report, a means for emotion analysis, and a means for adjusting answers based on emotions.

[1361] Hardware and Software Used

[1362] Hardware: Smartphone

[1363] Software: Python, OpenAI API, TextBlob

[1364] Data processing and calculation

[1365] An automated way to answer investor questions

[1366] The server receives questions entered by investors via their smartphones. The questions are analyzed using natural language processing technology and appropriate answers are generated. In this case, answers to questions are generated using the OpenAI API.

[1367] A means of presenting the summary of financial report documents and short reports, as well as points of concern

[1368] The server uses text mining technology to analyze financial reports and short messages and extract important information, which is then presented to investors as a summary.

[1369] Means of responding to key performance indicators other than financial statements

[1370] The server provides information on indicators that are not included in the financial statements but are important for evaluating the company's performance. This information is presented appropriately in response to investors' questions.

[1371] Emotion analysis means

[1372] The server uses TextBlob to analyze the sentiment of the investor's questions and comments, and based on the analysis, determines whether the question is positive or negative.

[1373] A way to tailor your answers based on emotions

[1374] The server generates an appropriate response according to the emotion of the investor based on the emotion information obtained by the emotion analysis means. For example, if the investor is feeling anxious, the server generates a response that gives the investor a sense of security.

[1375] Examples

[1376] Q: "What are some recent market trends?"

[1377] Answer: "The market has been very volatile recently, with technology stocks being particularly hard hit."

[1378] Examples of prompt statements

[1379] Answer investor questions: What have been the recent market trends?

[1380] In this way, the server can provide appropriate answers to investors' questions in real time and respond according to the investors' emotions.

[1381] The flow of the specific process in the application example 1 will be described with reference to FIG.

[1382] Step 1:

[1383] A user enters an investment question using a smartphone.

[1384] Input: The question text entered by the user.

[1385] Output: The question text sent to the server.

[1386] Step 2:

[1387] The server analyzes the received question text using natural language processing technology.

[1388] Input: The question text submitted by the user

[1389] Output: The parsed meaning or intent of the question

[1390] Step 3:

[1391] Based on the analysis results, the server generates an appropriate answer using the OpenAI API.

[1392] Input: The parsed meaning or intent of the question.

[1393] Output: The generated answer text

[1394] Step 4:

[1395] The server uses TextBlob to parse sentiment from the user's question text.

[1396] Input: The question text submitted by the user

[1397] Output: Analyzed sentiment information (positive, negative, etc.)

[1398] Step 5:

[1399] The server then tailors the generated answer based on the sentiment analysis results, for example generating a reassuring answer if the user is feeling anxious.

[1400] Input: Generated answer text, parsed sentiment information

[1401] Output: Answer text adjusted according to sentiment

[1402] Step 6:

[1403] The server analyzes financial report documents and short messages using text mining technology to extract important information.

[1404] Input: Text data of financial reports and short messages

[1405] Output: Summary of key information extracted

[1406] Step 7:

[1407] A server provides information regarding key performance indicators other than financial statements.

[1408] Input: The user's question text

[1409] Output: Key Performance Indicator Answer Text

[1410] Step 8:

[1411] The server sends the generated answer text to the user's smartphone.

[1412] Input: Adjusted answer text

[1413] Output: Answer text displayed on the user's smartphone

[1414] In this way, through the specific actions performed at each step, it is possible to provide appropriate answers to the user's questions and respond in accordance with the user's emotions.

[1415] Example 2

[1416] Next, a description will be given of Example 2 of the embodiment 2. In the following description, the data processing device 12 is referred to as a "server" and the smart glasses 214 are referred to as a "terminal".

[1417] Conventional investor information systems have difficulty quickly and accurately extracting important information from financial report materials and short messages and presenting it to investors in an appropriate format. In addition, they do not provide information based on investors' emotions, so they are unable to fully meet investors' needs. Furthermore, they lack the functionality to automatically generate appropriate answers to investors' questions.

[1418] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering indicators other than financial reports, a means for answering questions from investors, a means for recognizing investors' emotions using an emotion engine and presenting information according to the emotions, and a means for extracting important information from financial report materials and short messages using text mining technology. This makes it possible to provide investors with information quickly and accurately, and also makes it possible to present information according to investors' emotions. In addition, appropriate answers can be automatically generated to questions from investors.

[1419] An "investor" is an individual or entity that provides funds to a company or financial product in the hope of receiving a profit in return.

[1420] "Means for automatically answering questions" refers to technologies or systems that analyze questions from investors and generate appropriate answers.

[1421] A "financial report" is a document prepared by a company at regular intervals to report on its financial status and business performance.

[1422] A "tanshin" is a report that briefly summarizes a company's performance and financial status.

[1423] A "summary" is a concise summary that extracts important information from a long piece of text or report.

[1424] "Means of presenting sections of information of particular interest to investors" refers to technologies or systems for extracting and presenting information of particular interest to investors from financial report documents and short messages.

[1425] An "indicator" is a number or data used to evaluate a company's performance and financial situation.

[1426] An "emotion engine" is a technology or system that analyzes a user's emotions and provides information according to those emotions.

[1427] "Text mining technology" is a technology for extracting useful information from large amounts of text data.

[1428] "Means for presenting information" refers to technologies and systems for displaying extracted information to users in an easy-to-understand manner.

[1429] This invention relates to an information provision system for investors, and in particular to a system that extracts important information from financial report materials and short messages and presents it to investors in an appropriate format. Furthermore, it also has a function of providing information according to the sentiments of investors.

[1430] System Configuration

[1431] This system mainly consists of three elements: a server, a terminal, and a user.

[1432] server

[1433] The server has the following functions:

[1434] 1. Acquisition of financial report materials and short messages: The server acquires financial report materials and short messages from the official websites of companies and financial information providers. Specifically, it sends an HTTP request to download PDF files and saves them in local storage.

[1435] 2. Information extraction through text mining: The server converts the acquired materials into text format and extracts important information using natural language processing libraries "NLTK" and "SpaCy."

[1436] 3. Information summary generation: The server generates a summary based on the extracted information. The summary includes important indicators such as sales and profit margins. Specifically, the server uses the Gensim library to generate the summary text.

[1437] 4. Sentiment analysis using sentiment engine: The server uses IBM Watson Tone Analyzer and Microsoft Azure Text Analytics to analyze investor sentiment. It obtains investor feedback and comments and sends them to the sentiment engine.

[1438] 5. Emotion-based information display: The server displays appropriate information to investors based on the results of emotion analysis. For example, if an investor is excited, it will select positive information and display it through the web interface or mobile app.

[1439] Terminal

[1440] The terminal has the role of displaying the information received from the server to the user, specifically through a web browser or a mobile app.

[1441] User

[1442] The users are investors who use the information provided by the system. The users can ask the system questions, and the server automatically generates answers to those questions.

[1443] Examples

[1444] Example 1: Extract sales and profit margins from financial report documents and present summary information such as "Sales for the first quarter of fiscal year 2023 were 100 million yen, and profit margins were 15%."

[1445] Example 2: If the sentiment engine detects that investors are excited, it will present positive information such as "New product launch is expected to significantly increase sales."

[1446] Examples of prompt statements

[1447] "Extract key indicators from financial report documents and generate summary information."

[1448] "Analyze sentiment using a sentiment engine to present information according to investor sentiment."

[1449] In this way, the server is a system that uses text mining technology and an emotion engine to extract important information from financial report documents and short messages, and presents appropriate information to investors.

[1450] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1451] Step 1:

[1452] Acquisition of financial report materials and short reports

[1453] Input: The URL of the company's official website or financial information service.

[1454] Specific behavior: The server sends an HTTP request to download the PDF file from the specified URL. For example, the server retrieves the PDF file from the URL "https: / / example.com / financial_report.pdf".

[1455] Data processing: Save the downloaded PDF file to local storage.

[1456] Output: A PDF file saved in local storage.

[1457] Step 2:

[1458] Information Extraction through Text Mining

[1459] Input: A PDF file saved in local storage.

[1460] What it does: The server converts PDF files to text format using the PyPDF2 library, then analyzes the text using NLTK and SpaCy to extract key metrics like sales and profit margins.

[1461] Data processing: The data converted to text format is analyzed using natural language processing libraries to extract important information.

[1462] Output: Extracted key information (e.g. sales, profit margins).

[1463] Step 3:

[1464] Generate a summary of information

[1465] Input: Extracted key information.

[1466] Specific operation: The server uses the "Gensim" library to summarize the extracted information and generate a summary sentence, such as "Sales in the first quarter of fiscal year 2023 will be 100 million yen, and the profit margin will be 15%."

[1467] Data processing: Organize the extracted information and generate a summary.

[1468] Output: The generated summary.

[1469] Step 4:

[1470] Sentiment analysis with emotion engine

[1471] Input: Investor feedback and comments.

[1472] How it works: The server sends investor feedback and comments to IBM Watson Tone Analyzer and Microsoft Azure Text Analytics for sentiment analysis.

[1473] Data processing: Send feedback and comments to the sentiment engine and receive sentiment analysis results.

[1474] Output: Emotion analysis results (e.g. excitement, relief, anxiety).

[1475] Step 5:

[1476] Information presentation based on emotions

[1477] Input: Sentiment analysis results and generated summary sentences.

[1478] How it works: Based on the results of sentiment analysis, the server selects appropriate information for investors and displays it through a web interface or mobile app. For example, if an investor is excited, it will present positive information such as "Sales are expected to increase significantly due to the launch of a new product."

[1479] Data processing: Select information based on the results of sentiment analysis and summaries, and convert it into a display format.

[1480] Output: Information presented to investors.

[1481] In this way, the server processes and analyzes the input data at each step, and ultimately provides appropriate information to investors.

[1482] (Application example 2)

[1483] Next, a description will be given of Application Example 2 of Form 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".

[1484] In conventional investment support systems, it was difficult for investors to quickly extract and understand important information from financial reports and short messages. In addition, information was not presented according to investors' emotions, and appropriate information was not provided according to investors' needs. Furthermore, there was a lack of real-time information notifications, so investors were unable to receive important information in a timely manner.

[1485] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1486] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering indicators other than financial reports, a means for answering questions from investors, a means for recognizing investors' emotions and presenting information according to those emotions, and a means for notifying investors in real time. This enables investors to quickly understand important information and receive appropriate information according to their emotions. Also, by notifying information in real time, investors can receive important information in a timely manner.

[1487] "Means of automatically responding to questions from investors" refers to a function that analyzes questions entered by investors, generates appropriate answers, and automatically provides them.

[1488] "Means of presenting summaries of financial reports and short messages and points of concern" is a function that extracts important information from financial reports and short messages, summarizes it, and presents it to investors.

[1489] "Means of answering questions about indicators other than financial statements" is a function that provides answers to questions about important indicators that are not included in financial statements.

[1490] "Means of answering investor questions" is a function that provides appropriate answers to questions that investors have.

[1491] "Means of recognizing investor emotions and presenting information in accordance with those emotions" refers to a function that recognizes investor emotions using an emotion engine and provides information in accordance with those emotions.

[1492] "Means of notifying investors in real time" refers to a function that notifies investors of important information in real time.

[1493] The system for carrying out this invention is composed of a server and an investor's terminal. The server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering indicators other than financial reports, a means for answering questions from investors, a means for recognizing investor emotions and presenting information according to those emotions, and a means for notifying investors in real time.

[1494] Hardware and Software Used

[1495] Hardware: Servers, smartphones

[1496] Software: Python, requests library, textblob library, transformers library

[1497] Data processing and calculation

[1498] 1. Automated ways to answer investor questions:

[1499] The server uses natural language processing technology to analyze questions from investors and generate appropriate answers. Specifically, it uses a Python natural language processing library to analyze questions, retrieves relevant information from a database, and generates answers.

[1500] 2. How to present the summary of financial report documents and brief notes and points of concern:

[1501] The server uses text mining techniques to extract important information from financial reports and short reports, summarizes them, and presents them to investors. Specifically, it uses the transformers library to summarize the text.

[1502] 3. For non-financial report metrics, please provide:

[1503] The server provides answers to questions about important metrics that aren't included in financial reports, including pulling data from external databases and APIs.

[1504] 4. How investors can get answers to their questions:

[1505] The server provides appropriate answers to questions that investors have by analyzing the questions using natural language processing technology and providing relevant information.

[1506] 5. Ways to recognise investor sentiment and present information according to that sentiment:

[1507] The server uses the textblob library to recognize the sentiment of the investor and provide information according to that sentiment, for example, if the investor is excited, it provides positive information.

[1508] 6. Means of real-time investor notification:

[1509] The server notifies investors of important information in real time, including via push notifications and email notifications.

[1510] Examples

[1511] For example, if an investor types, "I'm very excited about the recent financial results!", the server will use the textblob library to recognize positive sentiment and provide positive information. It will also generate a summary of the earnings call and notify the investor in real time.

[1512] Examples of prompt statements

[1513] "Generate a summary of financial statements."

[1514] "Recognize investor sentiment and provide information that responds to that sentiment."

[1515] In this way, a system can be realized that helps investors obtain information more effectively.

[1516] The flow of the specific process in the application example 2 will be described with reference to FIG.

[1517] Step 1:

[1518] A user inputs a question using a smartphone. The input question is sent to a server. The input data is text in a natural language format.

[1519] Step 2:

[1520] The server analyzes user questions using natural language processing technology. Specifically, it uses a Python natural language processing library to tokenize the questions and perform semantic analysis. The input data is the user's question text, and the output data is the semantic information of the analyzed question.

[1521] Step 3:

[1522] The server retrieves relevant information from the database based on the semantic information of the parsed question. Specifically, it uses an SQL query to extract the required information from the database. The input data is the semantic information of the parsed question, and the output data is the information retrieved from the database.

[1523] Step 4:

[1524] The server generates an appropriate answer based on the acquired information. Specifically, it uses a generative AI model to convert the acquired information into an answer in natural language format. The input data is the information acquired from the database, and the output data is the generated answer text.

[1525] Step 5:

[1526] The server transmits the generated answer to the user's smartphone. The user can check the answer on the smartphone. The input data is the generated answer text, and the output data is the answer displayed on the user's smartphone.

[1527] Step 6:

[1528] The server retrieves financial reports and short messages and extracts important information using text mining techniques. Specifically, it summarizes the text using the transformers library. The input data is the text of financial reports and short messages, and the output data is the summarized important information.

[1529] Step 7:

[1530] The server uses the textblob library to recognize the user's emotion. Specifically, the server analyzes the user's input text and calculates an emotion score. The input data is the user's input text, and the output data is the emotion score.

[1531] Step 8:

[1532] The server presents emotion-specific information based on the recognized emotion score. For example, if a positive emotion is recognized, it provides positive information. The input data are the emotion score and the summarized important information, and the output data is emotion-specific information.

[1533] Step 9:

[1534] The server notifies the user of important information in real time. Specifically, the server sends the information to the user using push notifications or email notifications. The input data is the information according to the emotion, and the output data is the notification sent to the user.

[1535] Example 3

[1536] Next, a third embodiment of the third embodiment will be described. 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".

[1537] Conventional investor information systems have difficulty in providing appropriate answers to questions about key performance indicators (KPIs) that are not included in financial statements. In addition, they provide uniform answers without considering investors' emotions, which means that investors' doubts and concerns cannot be fully resolved. Furthermore, question analysis using natural language processing technology and emotion analysis using emotion recognition technology are insufficient, which can result in a decline in the quality of answers provided to investors.

[1538] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1539] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering key performance indicators other than the financial report, a means for analyzing investor emotions using emotion recognition technology and generating answers according to the emotions, and a means for analyzing questions from investors using natural language processing technology and generating appropriate answers. This makes it possible to provide appropriate answers to questions from investors using information other than the financial report and generate answers according to investor emotions.

[1540] An "investor" is an individual or legal entity that provides capital to a company or project and expects to receive a profit in return.

[1541] "Means of automatically answering questions" refers to a function in which the system receives questions from investors and automatically generates and provides appropriate answers.

[1542] An "financial report" is a formal document that reports a company's financial condition and performance, usually issued on a quarterly or annual basis.

[1543] A "tanshin" is a report that briefly summarizes a company's financial results, and is intended to provide quick information to investors and other interested parties.

[1544] An "abstract" is a concise summary of the main points of a longer document or report.

[1545] "Means of presenting sections of concern" is a function that extracts and presents important information and sections of interest to investors from financial report documents and short messages.

[1546] A "key performance indicator (KPI)" is a specific indicator used to evaluate a company's business performance, and includes sales, profit margins, customer satisfaction, etc.

[1547] "Emotion recognition technology" is a technology for analyzing and recognizing emotions from a user's input and actions.

[1548] "Natural language processing technology" is a technology that enables computers to understand and analyze human language, and is used to analyze the intent of questions.

[1549] "Analysis" is the process of examining data or information in detail to understand its structure and meaning.

[1550] The "means for generating an appropriate answer" is a function that automatically generates the most appropriate answer to the user's question based on the analysis results.

[1551] This invention is a system that automatically provides answers to questions from investors, and includes technology for generating summaries of financial report materials and short messages, answers to key performance indicators (KPIs) other than financial reports, and answers according to investor sentiment. Specific embodiments of this system are described below.

[1552] Hardware and Software Used

[1553] Servers: Database servers, Web servers

[1554] Software: Natural Language Processing (NLP) engine, emotion recognition engine, database management system (DBMS)

[1555] System Configuration

[1556] 1. The user (investor) uses a terminal to input a question.

[1557] 2. The terminal sends the entered question to the server.

[1558] 3. The server receives the question and analyzes it using an NLP engine.

[1559] 4. Based on the analysis results, the server obtains necessary data from sources other than financial reports (for example, the results of customer satisfaction surveys).

[1560] 5. The server uses an emotion recognition engine to analyze the user's emotions.

[1561] 6. Based on the emotion analysis results, the server generates an appropriate answer according to the user's emotions.

[1562] 7. The server sends the generated response to the terminal.

[1563] 8. The terminal displays the answer to the user.

[1564] Examples

[1565] For example, explain what happens if an investor asks, "What is the customer satisfaction rate?"

[1566] 1. A user types into a terminal, "What is your customer satisfaction rating?"

[1567] 2. The device sends this question to the server.

[1568] 3. The server uses an NLP engine to analyze the question and understands that you are seeking information about "customer satisfaction."

[1569] 4. The server retrieves the customer satisfaction survey results from the database.

[1570] 5. The server uses an emotion recognition engine to analyze whether the user is confused or not.

[1571] 6. If the server determines that the user is confused, it generates a clear, concise answer to clear up the confusion.

[1572] 7. The server generates an answer such as "According to the latest customer satisfaction survey, customer satisfaction is 85%" and sends it to the terminal.

[1573] 8. The terminal displays this answer to the user.

[1574] Examples of prompt statements

[1575] Below is an example of a prompt sentence to be input to the generative AI model.

[1576] Investors ask: "What is the customer satisfaction rate?"

[1577] The system responds: "According to our most recent customer satisfaction survey, customer satisfaction is 85%."

[1578] In this way, the system provides appropriate answers to the questions of the investors. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1579] Step 1:

[1580] The user inputs a question into the terminal.

[1581] Input: A user types "What is your customer satisfaction rating?" into an input field.

[1582] Output: The entered question is stored in the device's memory.

[1583] Specific operation: The user inputs a question using the device's keyboard or touch screen. The device displays the input in real time and the user confirms the question by pressing the Enter key or clicking the Send button.

[1584] Step 2:

[1585] The terminal sends a question to the server.

[1586] Input: The question entered by the user.

[1587] Output: The question is sent to the server as an HTTP request.

[1588] Specific operation: The terminal converts the entered question into JSON format and sends it as an HTTP POST request to the server's API endpoint.

[1589] Step 3:

[1590] The server receives the question and analyzes it using an NLP engine.

[1591] Input: A question sent from the device (e.g., "What is your customer satisfaction level?").

[1592] Output: Analysis of the question intent (e.g., asking for information about "customer satisfaction").

[1593] Specific operation: The server parses the received JSON data and uses an NLP engine (e.g., SpaCy or BERT) to analyze the intent of the question. As a result of the analysis, it extracts the keywords and intent of the question.

[1594] Step 4:

[1595] The server retrieves the required data.

[1596] Input: Analysis results from an NLP engine (e.g. "Customer Satisfaction").

[1597] Output: The results of a customer satisfaction survey retrieved from a database (e.g., "Customer satisfaction is 85%").

[1598] Specific operation: Based on the analysis results, the server queries the database management system (DBMS) for related data. For example, it retrieves the latest survey results on "customer satisfaction" from the database.

[1599] Step 5:

[1600] The server analyzes the user's emotions using an emotion recognition engine.

[1601] Input: A user question (e.g., "What is the customer satisfaction rate?").

[1602] Output: User sentiment analysis (e.g. "I'm confused").

[1603] Specific operation: The server uses an emotion recognition engine (for example, IBM Watson or Microsoft Azure's emotion analysis API) to analyze the emotion from the user's question. As a result of the analysis, it determines whether the user is confused.

[1604] Step 6:

[1605] The server generates a response according to the emotion.

[1606] Input: Sentiment analysis results (e.g. "I'm confused") and information obtained from a database (e.g. "Customer satisfaction is 85%").

[1607] Output: An appropriate answer depending on the user's sentiment (e.g. "According to our latest customer satisfaction survey, customer satisfaction is 85%").

[1608] Specific operation: The server generates an answer according to the user's emotions based on the results of emotion analysis. For example, if the server determines that the user is confused, it generates a simple and clear answer.

[1609] Step 7:

[1610] The server sends the response to the terminal.

[1611] Input: The generated answer (e.g. "According to our most recent customer satisfaction survey, customer satisfaction is 85%.").

[1612] Output: The answer is sent to the device as an HTTP response.

[1613] Specific operation: The server converts the generated answer into JSON format and sends it to the terminal as an HTTP response.

[1614] Step 8:

[1615] The terminal displays the answer to the user.

[1616] Input: The answer sent by the server (e.g. "According to our most recent customer satisfaction survey, customer satisfaction is 85%.").

[1617] Output: The answer that is displayed to the user.

[1618] Specific operation: The terminal analyzes the received JSON data and displays it on the screen in a format that is easy for the user to see. For example, it displays "According to the latest customer satisfaction survey, customer satisfaction is 85%."

[1619] (Application example 3)

[1620] Next, a description will be given of Application Example 3 of Form Example 3. 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".

[1621] Conventional investor information systems have difficulty in providing appropriate answers to questions about KPIs that are not included in financial statements. In addition, they provide uniform answers without considering investors' feelings, which means that investors' doubts and concerns cannot be fully resolved. Furthermore, they are not convenient because they do not take into account the use on mobile devices such as smartphones.

[1622] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means. In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering KPIs other than the financial report, a means for recognizing investor emotions using an emotion engine and providing an answer according to the emotion, and an application installed on a smartphone. This makes it possible to provide appropriate KPI information from information sources other than the financial report in response to investor questions and provide an answer according to the investor's emotions. In addition, since it can be used on a smartphone, convenience is improved.

[1623] An "investor" is an individual or legal entity that provides capital to a company or project with the aim of receiving a profit in return.

[1624] A "question" is an inquiry that an investor poses to the system to clarify the information or concerns they have.

[1625] "Automatic response means" refers to a function in which the system automatically generates and provides appropriate responses to questions from investors.

[1626] A "financial report" is an official document that reports a company's financial status and performance.

[1627] A "tanshin" is a document that provides a brief report of a company's quarterly performance.

[1628] A "summary" is information that concisely summarizes the main points of financial reports or short reports.

[1629] "Parts of concern" are those parts that contain information that investors are particularly interested in.

[1630] "Presentation means" refers to the functionality by which the system displays information to investors.

[1631] "KPI" stands for Key Performance Indicator and is an index used to measure a company's performance and degree of goal achievement.

[1632] An "emotion engine" is a technology that recognizes investor emotions and responds accordingly.

[1633] A "smartphone" is a mobile phone that can connect to the Internet and use applications.

[1634] An "application" is software installed on a smartphone to provide specific functions or services.

[1635] To implement this invention, the following system configuration is necessary. The system includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering KPIs other than financial reports, a means for recognizing investor emotions using an emotion engine and answering according to those emotions, and an application installed on a smartphone.

[1636] Hardware and Software Configuration

[1637] Hardware:

[1638] Smartphone

[1639] software:

[1640] Python

[1641] Transformers Library (Hugging Face)

[1642] Requests Library

[1643] Data processing and calculation flow

[1644] server:

[1645] 1. Receive questions from investors.

[1646] 2. The question content is analyzed using natural language processing technology.

[1647] 3. Obtain KPI information relevant to your question from sources other than financial statements.

[1648] 4. Recognize investor sentiment using a sentiment engine.

[1649] 5. Generate answers that respond to investor sentiment.

[1650] Device (smartphone):

[1651] 1. Investors enter their questions through the application.

[1652] 2. Receive the response from the server and display it to the investor.

[1653] Examples

[1654] For example, if an investor asks, "What is the customer satisfaction rate?", the server retrieves the results of a customer satisfaction survey and uses an emotion engine to recognize the investor's emotions. If the investor is determined to be confused, the server generates a clear and concise answer such as, "To resolve your confusion, we provide the following information: Customer satisfaction rate is 85%," and displays it on the smartphone.

[1655] Examples of prompt statements

[1656] Q: What is the customer satisfaction rate?

[1657] Emotion: Confusion

[1658] Answer: To clear up any confusion, I provide the following information: Our customer satisfaction rate is 85%.

[1659] In this way, it is possible to provide appropriate KPI information to investors' questions and respond according to their emotions. In addition, the system can be used on smartphones, improving convenience.

[1660] The flow of the specific process in the application example 3 will be described with reference to FIG.

[1661] Step 1:

[1662] A user starts the application on their smartphone and inputs a question. The input question is sent as text data from the device to the server.

[1663] Step 2:

[1664] The server analyzes the received question using natural language processing technology. Specifically, it understands the intent of the question and extracts related keywords. The results of this analysis are used in the next step.

[1665] Step 3:

[1666] Based on the analysis results, the server retrieves relevant KPI information from sources other than financial reports. For example, if the question is about customer satisfaction, it retrieves the results of a customer satisfaction survey. This information can be retrieved from a database or an external API.

[1667] Step 4:

[1668] The server recognizes the user's emotion using an emotion engine based on the acquired KPI information. The emotion engine receives the question text data as input and outputs the user's emotion (e.g., confusion, relief, excitement, etc.).

[1669] Step 5:

[1670] The server generates an appropriate answer depending on the perceived emotion. For example, if the user is perceived as confused, the server generates a clear and concise answer such as "To clear up your confusion, we provide the following information: Customer satisfaction is 85%."

[1671] Step 6:

[1672] The server sends the generated answer to the terminal. The terminal displays the received answer to the user. The user can check the answer on the screen of their smartphone.

[1673] In this way, it is possible to provide appropriate KPI information to investors' questions and respond emotionally.

[1674] 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 a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1675] 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> ) and other generation AIs. The data generation model 58 is obtained by making a neural network perform deep learning. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image is input. The data generation model 58 performs inference on the input inference data according to 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.

[1676] Another example of generative AI is Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1677] 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 smart glasses 214.

[1678] [Third embodiment]

[1679] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1680] 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.

[1681] 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 wide area network (WAN) and / or a local area network (LAN).

[1682] 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.

[1683] 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 the voice according to instructions from the processor 46.

[1684] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a 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 (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[1685] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54. The transmission and reception of various types of information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[1686] Fig. 6 shows an example of 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.

[1687] 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.

[1688] 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.

[1689] In the headset type terminal 314, 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.

[1690] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1691] "Example 1"

[1692] One embodiment of the present invention is a chatbot system for investors. The system receives questions from investors and analyzes the content of the questions using natural language processing technology. Based on the analysis results, the system generates an appropriate answer and provides it to the investor. For example, if an investor asks, "What are the main points of the latest financial report?", the system analyzes the financial report materials, extracts the main points, and answers the investor.

[1693] "Example 2"

[1694] Another embodiment of the present invention is a function for presenting summaries of financial report materials and short messages and points of concern. This function uses text mining technology to extract important information from financial report materials and short messages and present it to investors. For example, important indicators such as sales and profit margins listed in financial report materials are extracted and summarized information is provided to investors.

[1695] "Example 3"

[1696] Furthermore, another embodiment of the present invention has a function for answering KPIs other than financial reports. This function provides answers to questions from investors about KPIs that are not listed in financial reports. For example, if an investor asks, "What is the level of customer satisfaction?", the system obtains information from sources other than financial reports, such as the results of a customer satisfaction survey, and answers the investor.

[1697] The process flow of each embodiment will be described below.

[1698] "Example 1"

[1699] Step 1: The system receives the investor's question.

[1700] Step 2: The system uses natural language processing techniques to analyze the content of the question.

[1701] Step 3: Based on the analysis results, the system generates an appropriate answer.

[1702] Step 4: Provide the generated responses to investors.

[1703] "Example 2"

[1704] Step 1: The system retrieves financial report materials and short reports.

[1705] Step 2: Use text mining techniques to extract important information from the acquired materials.

[1706] Step 3: Summarize the extracted information and present it to investors.

[1707] "Example 3"

[1708] Step 1: The system receives the investor's question.

[1709] Step 2: The system obtains the required information from sources other than financial statements.

[1710] Step 3: Based on the information obtained, the system answers the investor's questions.

[1711] Example 1

[1712] Next, a description will be given of Example 1 of the embodiment 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".

[1713] Conventional investor relations systems had the problem that it was difficult to quickly and accurately respond to questions from investors. In addition, there was a lack of a way to efficiently extract information on financial reports and other indicators and provide it to investors. This meant that investors had to spend a lot of time and effort to obtain the information they needed.

[1714] 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.

[1715] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting a summary of financial report materials and important points, a means for answering indicators other than financial reports, a means for analyzing questions using natural language processing technology and generating appropriate answers, a means for generating answers using a generative AI model, a means for generating prompt sentences and inputting them into the generative AI model, and a means for providing the generated answers to investors, thereby enabling investors to quickly and accurately obtain necessary information.

[1716] "Investor" refers to an individual or entity that invests money in stocks, bonds, and other financial instruments in the financial markets.

[1717] "Means of automatically answering questions" refers to the function of receiving questions from investors and having the system automatically generate and provide appropriate answers.

[1718] "Financial report materials" refers to reports on financial status and performance that are issued periodically by a company.

[1719] "Means of presenting summaries and key points" refers to the function of extracting important information from financial report documents, summarizing it concisely and presenting it to investors.

[1720] An "indicator" refers to numerical data or statistical information used to evaluate a company's performance and financial situation.

[1721] "Natural language processing technology" refers to technology that enables computers to understand and analyze human language.

[1722] A "generative AI model" refers to a model for generating text data using artificial intelligence.

[1723] A "prompt sentence" refers to an instruction sentence to be input into a generative AI model.

[1724] "Analysis" refers to the process of breaking down input data and understanding its meaning and structure.

[1725] "Means for generating an answer" refers to the function of creating an appropriate answer based on the analysis results.

[1726] "Means of providing" refers to the functionality for communicating the generated answers to investors.

[1727] This invention relates to a chatbot system for investors, which provides quick and accurate answers to questions from investors. This system operates among three parties: a server, a terminal, and a user.

[1728] The server receives questions from investors and analyzes them using natural language processing technology, for example using a generative AI model such as OpenAI's GPT-4. The server extracts the intent and keywords of the question and obtains information to generate an appropriate answer.

[1729] The server generates an appropriate answer based on the analysis results. For example, the server refers to the latest financial report documents stored in the database and extracts the main points. In this case, the server inputs the following prompt sentences to the generative AI model:

[1730] Prompt: "What are the key takeaways from your latest earnings report?"

[1731] The generative AI model will generate an answer based on this prompt, for example in the following format:

[1732] Answer: "The main takeaways from our most recent financial report were that revenue increased 10% year over year and net income increased 15%. We also saw strong sales of our new products, which gives us good prospects for future growth."

[1733] The server transmits the generated answer to the terminal. The terminal receives the answer transmitted from the server and displays it to the user. The user can check the answer on the screen of the terminal.

[1734] The specific hardware configuration of this system is that the server is a computer equipped with a high-performance processor and large-capacity memory, and the terminal is a device operated by the user, such as a PC or smartphone. The software includes a natural language processing engine, generative AI model, and database management system that run on the server side.

[1735] As a concrete example, consider the case where a user inputs a question such as "What are the main points of the latest financial report?" The device sends this question to the server, which analyzes it using GPT-4. The server retrieves the latest financial report materials from the database and extracts the main points. The server inputs the prompt sentence into the generative AI model and sends the generated answer to the device. The device displays the answer to the user, allowing the user to quickly obtain the information they need.

[1736] Thus, the present invention provides a system that allows investors to quickly and accurately obtain the information they need.

[1737] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1738] Step 1:

[1739] The user enters a question.

[1740] A user inputs a question about an investment into a terminal (PC or smartphone). For example, the user inputs, "What are the main points of the latest financial report?" The input question is stored as text data on the terminal.

[1741] Step 2:

[1742] The terminal sends a question to the server.

[1743] The terminal transmits the question input by the user to the server. At this time, the terminal transmits text data to the server via the Internet. The input data is the question text of the user.

[1744] Step 3:

[1745] The server receives the query.

[1746] The server receives the question sent from the terminal. The received question is stored in the server as text data. The input data is the user's question text sent from the terminal.

[1747] Step 4:

[1748] The server parses the question.

[1749] The server analyzes the received question using natural language processing technology. For this analysis, it uses a generative AI model such as OpenAI's GPT-4. The server extracts the intent and keywords of the question and obtains information to generate an appropriate answer. The input data is the user's question text, and the output data is the analysis result.

[1750] Step 5:

[1751] The server generates an appropriate response.

[1752] The server generates an appropriate answer based on the analysis results. For example, the server refers to the latest financial report documents stored in the database and extracts the main points. In this case, the server inputs the following prompt sentences to the generative AI model:

[1753] Prompt: "What are the key takeaways from your latest earnings report?"

[1754] The generative AI model generates an answer based on this prompt. The input data is the prompt and the analysis result, and the output data is the generated answer.

[1755] Step 6:

[1756] The server sends the response to the terminal.

[1757] The server transmits the generated answer to the terminal. At this time, the server transmits the answer to the terminal as text data. The input data is the generated answer text, and the output data is the answer text transmitted to the terminal.

[1758] Step 7:

[1759] The terminal displays the answer to the user.

[1760] The terminal receives the answer sent from the server and displays it to the user. The user can check the answer on the screen of the terminal. The input data is the answer text sent from the server, and the output data is the answer text displayed to the user.

[1761] (Application example 1)

[1762] Next, a description will be given of Application Example 1 of Form 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".

[1763] Conventional chatbot systems for investors have the ability to automatically answer questions from investors, but they lack the ability to provide appropriate answers to security-related questions.In addition, they have issues with not improving the user experience because they do not provide a user interface or retrieve information from a database efficiently.

[1764] 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.

[1765] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering performance indicators other than financial reports, a means for answering investors' concerns, a means for generating appropriate answers to security-related questions, a means for analyzing questions and generating appropriate answers using natural language processing technology, a means for acquiring information from a database and providing it to users, and a means for providing a user interface. This makes it possible to provide quick and accurate answers to questions from investors and provide comprehensive support including security-related information.

[1766] "Means of automatically responding to questions from investors" refers to a function that analyzes questions entered by investors and automatically generates and provides appropriate answers.

[1767] The "means of presenting summaries of financial report documents and short messages as well as points of interest" is a function that extracts important information from financial report documents and short messages and presents the user with summaries and points of particular interest.

[1768] The "means of answering questions about performance indicators other than financial statements" is a function that provides appropriate answers to questions about performance indicators other than financial statements.

[1769] "Means of answering investor questions" is a function that provides appropriate answers to questions that investors have.

[1770] The "means for generating appropriate answers to security-related questions" is a function for analyzing security questions and generating and providing appropriate answers.

[1771] "Means of analyzing questions using natural language processing technology and generating appropriate answers" refers to a function that uses natural language processing technology to analyze a user's question and generate an appropriate answer based on the content of the question.

[1772] "Means for retrieving information from a database and providing it to a user" refers to a function for searching information stored in a database, retrieving the necessary information, and providing it to a user.

[1773] "Means for providing a user interface" refers to a function that provides an interface for a user to input a question and receive a response.

[1774] A system for implementing this invention includes means for automatically answering questions from investors, means for presenting summaries of financial report materials and short messages and points of concern, means for answering performance indicators other than financial reports, means for answering investor concerns, means for generating appropriate answers to security-related questions, means for analyzing questions and generating appropriate answers using natural language processing technology, means for retrieving information from a database and providing it to a user, and means for providing a user interface.

[1775] System Configuration

[1776] The server includes a question analysis engine using natural language processing technology, a database access module, an answer generation engine, and a user interface module. These modules are implemented using Python and natural language processing libraries (e.g., spaCy and Transformers).

[1777] Program processing explanation

[1778] When the server receives a question from a user, it first analyzes the question using natural language processing technology. Natural language processing libraries such as spaCy and Transformers are used for this analysis. The analyzed question is used to retrieve related information from a database via a database access module. The database contains financial report documents, performance indicators, security-related information, and so on.

[1779] The acquired information is processed into appropriate answers by the answer generation engine, which uses a generative AI model to generate answers, and the generated answers are provided to the user through a user interface module.

[1780] Examples

[1781] For example, if a user asks, "What are the latest security threats?" the system would do the following:

[1782] 1. The user interface module receives a question.

[1783] 2. Analyze the question using natural language processing techniques.

[1784] 3. The database access module retrieves the relevant security information from the database.

[1785] 4. The answer generation engine generates appropriate answers based on the information obtained.

[1786] 5. The user interface module provides the generated answers to the user.

[1787] Examples of prompt statements

[1788] User: What are the latest security threats?

[1789] Systems: The latest security threat is ransomware attacks.

[1790] In this way, the system can provide quick and accurate answers to questions from investors and users, providing comprehensive support, including security-related information.

[1791] The flow of the specific process in the application example 1 will be described with reference to FIG.

[1792] Step 1:

[1793] The user inputs a question. The user inputs a question through the user interface of a smartphone or a robot. The input question is sent to the server.

[1794] Step 2:

[1795] The server receives the question. The server receives the question sent by the user and analyzes the question using natural language processing technology. Specifically, it uses natural language processing libraries such as spaCy and Transformers to analyze the meaning of the question. The input is the user's question, and the output is the structure of the analyzed question.

[1796] Step 3:

[1797] The server retrieves information from the database. Based on the content of the parsed question, the server searches the database for relevant information. The database may contain financial report documents, performance indicators, security-related information, etc. The input is the parsed question structure, and the output is the relevant information.

[1798] Step 4:

[1799] The server generates an answer. The server uses a generative AI model to generate an appropriate answer based on the acquired information. Specifically, the acquired information is input as a prompt into the generative AI model to generate an appropriate answer. The input is the information acquired from the database, and the output is the generated answer.

[1800] Step 5:

[1801] The server provides the answer to the user. The server provides the generated answer to the user through a user interface module. The user can check the answer on the screen of a smartphone or a robot. The input is the generated answer, and the output is the answer provided to the user.

[1802] Specifically, when a user asks "What are the latest security threats?", the server receives the question, analyzes it using natural language processing technology, and retrieves information about the latest security threats from a database. It then uses a generative AI model to generate the answer "The latest security threat is a ransomware attack" and provides it to the user.

[1803] Example 2

[1804] Next, a second embodiment of the second form will be described. 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".

[1805] In the conventional system, investors had to manually extract and summarize important information from financial reports and short messages, which was cumbersome and time-consuming. It was also difficult to quickly and appropriately respond to questions from investors, so an efficient means of resolving investors' concerns was needed.

[1806] 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.

[1807] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering questions about indicators other than financial reports, a means for uploading financial report materials and short messages, a means for receiving uploaded materials, a means for analyzing materials using a text mining tool, a means for extracting important indicators, a means for summarizing the extracted information, a means for formatting the summary in natural language using a generative AI model, a means for transmitting the formatted summary information to a terminal, and a means for the terminal to present the summary information to a user. This makes it possible to quickly and accurately provide information required by investors and efficiently resolve investors' questions.

[1808] An "investor" is an individual or legal entity that provides capital to a company or project and expects to receive a profit in return.

[1809] "Means for automatically answering questions" refers to a part of a system that has the function of receiving questions from investors, analyzing them using natural language processing technology, and generating appropriate answers.

[1810] A "financial report" is a document prepared by a company at regular intervals to report on its financial status and business performance.

[1811] A "tanshin" is a report that briefly summarizes a company's financial results.

[1812] A summary is a short sentence that concisely summarizes a longer piece of text or complex information.

[1813] "Means of presenting sections of points of concern" refers to part of a system that has the function of extracting and presenting important information and sections of interest to investors from financial report documents and short messages.

[1814] An "indicator" is a number or data used to evaluate a company's performance and financial situation.

[1815] An "uploading means" is a part of the system that has the functionality to allow a user to send files to the system.

[1816] The "receiving means" is a part of the system in which the server has the function of receiving and storing files sent by users.

[1817] A "text mining tool" is software for analyzing text data and extracting important information.

[1818] "Means for analysis" refers to a part of a system that has the function of analyzing the contents of documents using text mining tools.

[1819] The "means for extracting important indicators" is a part of the system that has the function of extracting important data such as sales revenue and profit margins from the analysis results.

[1820] The "summarizing means" is the part of the system that has the functionality to generate a concise summary based on the extracted key information.

[1821] A "generative AI model" is a model for generating sentences in natural language using artificial intelligence technology.

[1822] A "means for formatting a summary in natural language" is a part of a system that has the function of formatting summary information in easy-to-understand natural language using a generative AI model.

[1823] A "terminal" is a device through which a user accesses the system and receives information.

[1824] A "means for presenting" is a part of the system where a terminal has the ability to display information to a user.

[1825] This invention is a system that allows investors to quickly and accurately extract important information from financial reports and short messages, and provides summary information. The system operates in cooperation with a server, a terminal, and a user.

[1826] Hardware and Software Used

[1827] Hardware: Servers, terminals

[1828] Software: Text mining tools (e.g. NLTK, SpaCy), generative AI models (e.g. GPT-4)

[1829] System Operation Overview

[1830] 1. The user uploads financial report documents and short messages to the system.

[1831] 2. The server receives the uploaded materials and analyzes the text data using text mining tools.

[1832] 3. The server extracts key indicators such as sales revenue and profit margins from the analysis results.

[1833] 4. The server summarizes the extracted information and uses a generative AI model to format it in natural language for investors to understand.

[1834] 5. The server sends the formatted summary information to the terminal and presents it to the user.

[1835] Examples

[1836] For example, if a user uploads "Financial report materials for Q1 2023" to the system, the server will process it as follows:

[1837] 1. The user uploads the "2023 Q1 Financial Report Materials" to the system.

[1838] 2. The server receives the material and analyzes the text data using NLTK.

[1839] 3. The server extracts key indicators such as sales revenue and profit margins.

[1840] 4. The server summarizes the extracted information and uses GPT-4 to generate a natural language summary such as, "Revenue in Q1 2023 increased 10% compared to the same period last year, and profit margins reached 15%."

[1841] 5. The server sends the generated summary information to the terminal and presents it to the user.

[1842] Examples of prompt statements

[1843] An example of a prompt to be input to the generative AI model is as follows:

[1844] "Please extract and summarize key indicators from the text below. Text: 'In the financial results report for Q1 2023, sales increased 10% year-on-year and profit margins reached 15%. In addition, sales of new products are strong, and future growth is expected.'"

[1845] Using this prompt, the generative AI model extracts key information and generates a summary.

[1846] This system allows investors to quickly obtain the information they need, enabling them to make efficient investment decisions. It also allows for quick and appropriate responses to investor questions, resolving investors' doubts and providing highly reliable information.

[1847] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1848] Step 1:

[1849] The user uploads financial report documents and short messages to the system.

[1850] Specific operation: The user selects "Financial report materials for Q1 2023.pdf" and clicks the upload button.

[1851] Input: The financial report file selected by the user.

[1852] Output: The uploaded file is sent to the server.

[1853] Step 2:

[1854] The server receives the uploaded material.

[1855] Specific operation: The server receives "2023Q1Financial Report Materials.pdf" and saves it in the / uploads directory.

[1856] Input: Financial report file submitted by user.

[1857] Output: The file saved in the specified directory on the server.

[1858] Step 3:

[1859] The server analyzes the material using text mining tools.

[1860] Specific operation: The server uses NLTK to extract text from "2023Q1Financial Report Materials.pdf" and begins analysis.

[1861] Input: Financial report file stored on the server.

[1862] Output: The extracted text data.

[1863] Step 4:

[1864] The server extracts key indicators.

[1865] Specific operation: The server extracts important indicators based on keywords such as "sales revenue," "profit margin," and "year-on-year comparison."

[1866] Input: The extracted text data.

[1867] Output: Extracted key indicators data.

[1868] Step 5:

[1869] The server summarizes the extracted information.

[1870] Specific operation: The server generates summary information based on the extracted important indicators.

[1871] Input: Extracted key indicators data.

[1872] Output: Summarized information.

[1873] Step 6:

[1874] The server uses a generative AI model to format the summary in natural language.

[1875] How it works: The server inputs a prompt sentence into GPT-4, which generates a natural language summary such as, "Sales in Q1 2023 increased 10% year-on-year, and profit margins reached 15%."

[1876] Input: The summarized information and the prompt.

[1877] Output: Natural language formatted summary information.

[1878] Step 7:

[1879] The server transmits the formatted summary information to the terminal.

[1880] Specific operation: The server sends the generated summary information to the user's terminal as an HTTP response.

[1881] Input: A formatted summary in natural language.

[1882] Output: Summary information sent to the user's terminal.

[1883] Step 8:

[1884] The terminal presents the summary information to the user.

[1885] Specific behavior: The device displays on the browser, "Revenue in the first quarter of 2023 increased 10% year-on-year, and profit margin reached 15%."

[1886] Input: Summary information sent by the server.

[1887] Output: Summary information displayed to the user.

[1888] (Application example 2)

[1889] Next, a description will be given of Application Example 2 of Form 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".

[1890] Conventional investor information systems have difficulty quickly and accurately extracting and summarizing important information from financial reports and short messages. They also lack the ability to automatically generate appropriate answers to questions from investors. This means that investors have to spend a lot of time and effort to obtain the information they need.

[1891] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means. In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering indicators other than the financial report, and a means for extracting important information from the financial report materials and generating summaries through an application installed on a smartphone. This enables investors to quickly and accurately obtain the information they need.

[1892] An "investor" is an individual or legal entity that provides capital to a company or project with the aim of receiving a profit in return.

[1893] "Means for automatically answering questions" refers to technological means for analyzing questions from investors and generating appropriate answers.

[1894] A "financial report" is an official document that reports a company's financial status and performance.

[1895] A "tanshin" is a document that provides a concise report of a company's quarterly performance and financial status.

[1896] A "summary" is a short summary of a longer document or information, extracting and concisely presenting the important points.

[1897] "Means of presenting the sections of information of particular interest to investors" refers to technological means for extracting and presenting information of particular interest to investors from financial report documents and short messages.

[1898] An "indicator" is a specific number or data used to evaluate a company's performance and financial situation.

[1899] A "smartphone" is a mobile phone that can connect to the Internet and install applications.

[1900] An "Application" is a software program designed to provide a particular function or service.

[1901] "Text mining technology" is a data analysis technique for extracting useful information from large amounts of text data.

[1902] "Natural language processing technology" is a technology that enables computers to understand and analyze human language.

[1903] "Material information" refers to key data and indicators contained in financial reports and short reports that investors need to make decisions.

[1904] A "means for generating a summary" is a technical means for extracting important information from a long document and summarizing it concisely.

[1905] A system for implementing the present invention includes a server, a smartphone, and related software components. A specific configuration and operation of the system will be described below.

[1906] System Configuration

[1907] 1. Server:

[1908] The server uses natural language processing and text mining technologies to analyze questions from investors and generate appropriate answers.

[1909] The server extracts important information from financial reports and short messages and generates summaries.

[1910] The server then sends the generated summary and answers to the smartphone.

[1911] 2. Smartphone:

[1912] The smartphone displays the summary and answers sent from the server through a dedicated application.

[1913] A user can use a smartphone to input a question and send it to the server.

[1914] 3. Software Components:

[1915] A natural language processing library (e.g. spaCy): Used to parse investor questions and generate appropriate answers.

[1916] Text mining libraries (e.g. scikit-learn): Used to extract key information from financial reports and short reports.

[1917] Data processing and calculation

[1918] 1. Question analysis and answer generation:

[1919] The server receives questions from investors and analyzes them using natural language processing techniques.

[1920] Based on the parsed question, an appropriate answer is generated.

[1921] The generated answers are sent to the smartphone and displayed to the user.

[1922] 2. Generate summary of financial report documents:

[1923] The server receives financial report documents and short messages and uses text mining technology to extract important information.

[1924] A summary is generated based on the extracted information.

[1925] The generated summary is sent to the smartphone and displayed to the user.

[1926] Examples

[1927] Text from the financial report: "Financial report for fiscal year 2023. Sales increased 10% year-on-year, and profit margins improved 5%. Sales of new products were strong, with growth in the Asian market being particularly notable."

[1928] Example prompt: "Please extract and summarise the key information from this financial statement."

[1929] This system allows investors to quickly and accurately obtain the information they need, improving the efficiency of their decision-making. In addition, by using a smartphone, investors can enjoy the convenience of being able to access information anytime, anywhere.

[1930] The flow of the specific process in the application example 2 will be described with reference to FIG.

[1931] Step 1:

[1932] A user inputs a question using a smartphone and sends it to the server. The input question is sent to the server as text data.

[1933] Step 2:

[1934] The server analyzes the received question using natural language processing technology (e.g., spaCy). Specifically, it analyzes the grammatical structure of the question and extracts important keywords and phrases. The input is the user's question text, and the output is the analyzed keywords and phrases.

[1935] Step 3:

[1936] The server generates appropriate answers based on the analyzed keywords and phrases. It uses a generative AI model to generate the best answer to the question. The input is the analyzed keywords and phrases, and the output is the generated answer text.

[1937] Step 4:

[1938] The server sends the generated answer to the smartphone, which displays the received answer to the user. The input is the generated answer text, and the output is the answer displayed on the smartphone screen.

[1939] Step 5:

[1940] A user uploads financial report documents using a smartphone. The uploaded documents are sent to the server as text data.

[1941] Step 6:

[1942] The server analyzes the received financial report documents using text mining techniques (e.g., scikit-learn). Specifically, it extracts important information from the documents. The input is the text of the financial report documents, and the output is the extracted important information.

[1943] Step 7:

[1944] The server generates a summary based on the extracted key information. A generative AI model is used to generate a summary of the material. The input is the extracted key information, and the output is the generated summary text.

[1945] Step 8:

[1946] The server sends the generated summary to the smartphone, which displays the received summary to the user. The input is the generated summary text, and the output is the summary displayed on the smartphone screen.

[1947] In this way, users can enter questions and upload financial report documents via their smartphones and receive quick and accurate answers and summaries from the server.

[1948] Example 3

[1949] Next, a third embodiment of the third embodiment will be described. 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".

[1950] In the conventional system, it was difficult to answer questions from investors about indicators that were not included in financial reports, and the process of identifying appropriate sources, acquiring data, and generating answers using generative AI models was not automated, making it impossible to provide quick and accurate answers. This resulted in a decrease in investor satisfaction with their questions.

[1951] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1952] In this invention, the server includes a means for automatically answering questions from investors, a means for presenting summaries of financial report materials and short messages and points of concern, a means for answering indicators other than financial reports, a means for answering questions from investors, a means for analyzing questions using natural language processing technology and identifying appropriate information sources, a means for acquiring data from information sources and generating answers using a generative AI model, and a means for providing the generated answers to investors, thereby making it possible to provide quick and accurate answers to questions from investors.

[1953] "Means of automatically responding to questions from investors" refers to a function whereby the system automatically generates and provides appropriate answers to questions entered by investors.

[1954] The "means of presenting summaries of financial reports and short messages, as well as points of concern" is a function that extracts important information from financial reports and short messages, and presents summaries and points of interest to investors.

[1955] "Means of answering questions about indicators other than financial statements" refers to a function that obtains information about indicators not listed in financial statements and provides answers to investors' questions.

[1956] "Means of answering investor questions" refers to the function whereby the system provides appropriate answers to questions that investors have.

[1957] "Means for analyzing questions using natural language processing technology and identifying appropriate information sources" is a function for analyzing questions from investors using natural language processing technology and identifying appropriate information sources for those questions.

[1958] "Means of obtaining data from an information source and generating an answer using a generative AI model" refers to a function that obtains necessary data from an identified information source and generates an answer using a generative AI model based on that data.

[1959] "Means of providing generated answers to investors" refers to a function that provides answers generated by a generative AI model to investors.

[1960] This invention relates to a system that automatically provides answers to questions from investors. The system operates in cooperation with a server, a terminal, and a user.

[1961] The server generates a program using the generative AI model. This program has the function of providing answers to questions from investors, even for indicators not included in financial reports. Specifically, the server analyzes questions using natural language processing technology, identifies appropriate sources, obtains data from the sources, and generates answers using the generative AI model.

[1962] In terms of hardware, the server is a computer system equipped with a high-performance processor and large memory capacity. In terms of software, libraries such as spaCy and NLTK are used for natural language processing technology, and OpenAI's GPT-4 is used for generative AI models. MySQL, PostgreSQL, etc. are used as database management systems (DBMS).

[1963] The terminal provides an interface for the user to input a question. The terminal communicates with the server through a web browser or a dedicated application, sends the question, and receives and displays the answer from the server.

[1964] A user (investor) inputs a question through the terminal interface. For example, "What is the level of customer satisfaction?" The terminal sends this question to the server. The server analyzes the question, identifies appropriate sources, retrieves data from the sources, and generates an answer using a generative AI model. The generated answer is provided to the user through the terminal.

[1965] As a concrete example, consider a case where an investor asks, "What is the level of customer satisfaction?" When this question is sent from the terminal to the server, the server retrieves the results of the customer satisfaction survey and uses a generative AI model to generate an answer like the following:

[1966] Example prompt:

[1967] Investors are asking: "What is the customer satisfaction rate?"

[1968] Server action:

[1969] 1. Analyze the questions to identify information about customer satisfaction.

[1970] 2. Obtain the results of the customer satisfaction survey via DBMS or API.

[1971] 3. Use a generative AI model (e.g. GPT-4) to generate an answer based on the acquired data.

[1972] 4. Provide the generated answers to the investors via the terminal.

[1973] Example of a generated answer:

[1974] "According to our most recent customer satisfaction survey, our customer satisfaction rate is 85%. The survey was conducted in March 2023 and covered 1,000 customers."

[1975] In this way, a system is realized in which the server, the terminals, and the users work together to provide prompt and accurate answers to questions from investors. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1976] Step 1:

[1977] The user enters a question.

[1978] A user (investor) inputs a question through the terminal interface. For example, the user inputs "What is the level of customer satisfaction?" The input question is temporarily stored in the terminal memory.

[1979] Step 2:

[1980] The terminal sends a question to the server.

[1981] The terminal sends the entered question to the server as an HTTP POST request. The request contains the question in JSON format. The input is the user's question, and the output is the request sent to the server.

[1982] Step 3:

[1983] The server parses the question.

[1984] The server analyzes the received HTTP request and extracts the question content. It uses a natural language processing (NLP) library (e.g., spaCy or NLTK) to analyze the intent of the question and identify the keyword "customer satisfaction." The input is the question data from the terminal, and the output is the analyzed keyword.

[1985] Step 4:

[1986] The server identifies the appropriate source of information.

[1987] The server identifies a database or external API for customer satisfaction surveys based on the analysis results. For example, it selects a table in a MySQL database where the results of the customer satisfaction surveys are stored. The input is the analyzed keywords, and the output is the identified information source.

[1988] Step 5:

[1989] The server retrieves the data from the information source.

[1990] The server retrieves data from the identified sources. For example, it might run a SQL query against a MySQL database to retrieve customer satisfaction survey results, or, in the case of an external API, it might send an HTTP GET request to an API endpoint to retrieve the data. The input is the identified sources and the output is the retrieved data.

[1991] Step 6:

[1992] The server generates an answer using a generative AI model.

[1993] The server inputs the acquired data into a generative AI model (e.g., GPT-4) to generate an answer in natural language. The generative AI model generates an answer based on the input data, taking into account appropriate context. The input is the acquired data, and the output is the generated answer.

[1994] Step 7:

[1995] The server sends the generated answer to the terminal.

[1996] The server sends the generated answer to the terminal as an HTTP response. The response includes the generated answer in JSON format. The input is the generated answer, and the output is the response sent to the terminal.

[1997] Step 8:

[1998] The terminal displays the answer to the user.

[1999] The terminal analyzes the received HTTP response and displays the response content on the user interface. The user can check the respon...

Claims

1. A means to automatically answer questions from investors; A means of presenting summaries of financial report materials and short reports, as well as points of concern, Means of responding to indicators other than the above financial report materials and short reports, A means for answering questions from the investor; means for recognizing sentiment of said investor using a sentiment engine and presenting information according to said sentiment; A means for extracting important information from the financial report materials and short messages using text mining technology; A system including:

2. The system of claim 1 , wherein the means for automatically answering investor questions further comprises means for analyzing investor questions using natural language processing techniques and generating appropriate answers.

3. The system of claim 1, wherein the means for presenting a summary of the financial report materials / short notes and points of concern further includes means for extracting important information from the financial report materials / short notes using text mining technology and presenting the information to investors.

Images