system

Abstract

Provide a system. A means for periodically collecting news information; A means for analyzing the collected news information using natural language processing; A means for extracting important data from the analyzed information; a means for generating a summary based on the extracted information; a means for storing the generated summaries in a database; means for providing the stored summary to a user terminal; A system including:

Description

[Technical Field]

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

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

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

[0004] In recent years, investors need to quickly obtain important information from the vast amount of news. However, conventional methods require a great deal of time and effort to collect investment-related news, extract important information, and summarize it, making it difficult to obtain information efficiently. Furthermore, there is no system in place to immediately use the information obtained, raising concerns that the time lag in providing information could have a negative impact on investment decisions. [Means for solving the problem]

[0005] The system of the present invention includes a means for periodically collecting news information, a means for analyzing the collected news information using natural language processing, a means for extracting important data from the analyzed information, a means for generating summaries based on the extracted information, a means for saving the generated summaries in a database, and a means for providing the saved summaries to a user terminal. This system enables investors to quickly obtain important information from a vast amount of news and use it immediately. News information is obtained from multiple news sources, and summaries are efficiently generated using a text summarization algorithm, minimizing the time lag in providing information.

[0006] "News information" refers to articles and data reporting investment-related events and market trends.

[0007] "Means of collection" refers to the functions and mechanisms for periodically obtaining data from specified news sources from the Internet.

[0008] "Natural language processing" refers to the technology that enables computers to analyze, understand, and manipulate human language.

[0009] "Means of analysis" refers to the functions and mechanisms for analyzing news information obtained using natural language processing technology and extracting important elements.

[0010] "Important data" refers to data that specifically indicates information necessary for investment decisions, such as factors that affect stock prices and market trends.

[0011] "Means of extraction" refers to the functions and mechanisms for extracting important data from news articles.

[0012] The "means for generating a summary" refers to a function or mechanism for creating a concise summary based on the extracted information in a form that is easy for investors to understand.

[0013] "Database" means an information repository for storing collected, analyzed, and summarized data. This includes SQL and NoSQL databases.

[0014] "Storage means" refers to the functions and mechanisms for recording the generated summaries in a database.

[0015] "User Device" means a device, such as a computer, tablet, or smartphone, that a User uses to receive information.

[0016] "Means for providing" refers to the functions and mechanisms for transmitting the summaries stored in the database to the user's terminal and displaying them. [Brief explanation of the drawings]

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

[0018] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0020] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0021] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0022] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0023] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0025] [First embodiment]

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

[0027] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0030] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0034] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0037] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0038] The system of the present invention collects, analyzes, summarizes, and provides investment-related news quickly and efficiently. To implement this system, the following specific forms are considered and implemented.

[0039] Gathering news information

[0040] News gathering methods

[0041] The server pre-configures the URLs and API endpoints of specific news sources (e.g., major financial news sites or data providers) and periodically collects news data based on these. For example, the server retrieves the latest news from the Wall Street Journal or Reuters APIs at 8:00 every day.

[0042] Example: The server accesses the API endpoint "https: / / example-news-api.com / v2 / latest" to retrieve the latest news data in JSON format.

[0043] News information analysis

[0044] Natural language processing tools

[0045] The server analyzes the collected news data using natural language processing (NLP) techniques. NLP libraries (e.g., spaCy and NLTK) are used to analyze the content of news articles and identify important keywords and sentence structures.

[0046] Example: The server analyzes the news article "Nikkei average stock price rises 2%" and extracts keywords such as "Nikkei average," "stock price," and "2% rise."

[0047] Extraction of important information

[0048] Important information extraction means

[0049] Based on the analysis results, the server filters and extracts important information necessary for investment decisions (for example, causes of stock price fluctuations, corporate earnings announcements, etc.).

[0050] Example: From the news data analyzed by the server, specific information such as "The main reason for the 2% rise in the Nikkei average stock price is the improved performance of export companies" is extracted.

[0051] Generate a summary

[0052] Summary generation means

[0053] Based on the key information extracted by the server, a summarization algorithm is used to generate a summary that is easy for investors to understand. Abstractive summarization and extractive summarization techniques can be combined.

[0054] Example: The server generates a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0055] Save Summary

[0056] Database storage means

[0057] The server stores the generated summaries in a database, including the summary content, a link to the original news article, and the date and time the summary was generated.

[0058] Example: The server stores in an SQL database the summary "The Nikkei average rose 2%, primarily due to improved performance of export companies" along with further information.

[0059] Provision to users

[0060] User terminal provision means

[0061] When a user accesses the Yahoo Finance portal, the server retrieves the latest summary from the database and provides it to the user's device, allowing the user to make investment decisions based on it.

[0062] Example: When a user visits the "Latest News" section of Yahoo Finance, they are presented with a server-provided summary that reads, "Nikkei average rises 2%, primarily due to improved performance of export companies."

[0063] The system of the present invention includes these means and is capable of collecting, analyzing, summarizing and providing investment-related news quickly and efficiently, thereby greatly assisting investors in obtaining information.

[0064] The processing flow will be explained below.

[0065] Specific processing steps of the news gathering and analysis system

[0066] Step 1:

[0067] The server configures the URL and API endpoint of the news source, for example, registering the endpoint of a particular financial news site or data provider.

[0068] Step 2:

[0069] The server retrieves news data from multiple news sources according to a pre-defined schedule, for example, calling the API every day at 8:00 to retrieve the latest news.

[0070] Step 3:

[0071] The news data retrieved by the server is preprocessed, specifically by extracting text from HTML content and converting it into a parseable format.

[0072] Step 4:

[0073] The server uses natural language processing (NLP) techniques to analyze the preprocessed news data, using NLP libraries to analyze sentence structure and extract keywords.

[0074] Step 5:

[0075] The server then uses the analysis results to extract key information from news articles, filtering the information based on specific keywords such as "stock prices," "earnings announcements," and "market trends."

[0076] Step 6:

[0077] The server organizes the extracted key information and applies a summary generation algorithm to generate summaries that are easy for investors to understand. Abstractive summaries and extractive summaries are used.

[0078] Step 7:

[0079] The server stores the generated summary in a database, including the summary content, the URL of the original article, and the date and time the summary was generated.

[0080] Step 8:

[0081] A user visits the Yahoo! Finance portal and requests to view breaking news.

[0082] Step 9:

[0083] The server retrieves the latest summary data from the database and sends it to the user terminal.

[0084] Step 10:

[0085] The user's device displays the received summary data on the Yahoo Finance interface, allowing the user to view the information.

[0086] Example 1

[0087] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0088] Because news information is vast and scattered, investors do not have time to research and analyze it individually, making it difficult to quickly and efficiently obtain important news information. Furthermore, analyzing news information and generating summaries requires advanced technology and knowledge, and doing this manually is labor-intensive. Therefore, there is a need for a system that can quickly and accurately provide the important news information investors need.

[0089] 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.

[0090] In this invention, the server includes means for periodically collecting news information, means for preprocessing the collected news information, means for analyzing the preprocessed news information using natural language processing, means for extracting important data from the analyzed information, means for generating summaries based on the extracted important data, means for storing the generated summaries in a database, and means for providing the stored summaries to a user terminal, thereby enabling investors to quickly and efficiently obtain the latest important news information and make investment decisions.

[0091] "Means for regularly collecting news information" refers to a function that allows you to set the URL or API endpoint of a specific news source and obtain news data daily or at a fixed time.

[0092] "Means for preprocessing collected news information" refers to a function for removing unnecessary information from acquired news data and converting it into a format that can be analyzed.

[0093] "Means for analyzing preprocessed news information using natural language processing" refers to a function that uses natural language processing techniques and libraries to analyze preprocessed news text and identify important keywords and sentence structures.

[0094] "Means of extracting important data from analyzed information" refers to a function that filters and selects important information necessary for investment decisions from the data obtained as a result of natural language processing.

[0095] The "means for generating a summary based on the extracted important data" is a function that uses an algorithm to create a summary based on the extracted important information in a form that is easy for investors to understand.

[0096] "Means for saving the generated summary in a database" refers to a function that saves the generated summary and its related information in a database so that it can be referenced or searched later.

[0097] The "means for providing the stored summary to the user terminal" is a function for retrieving the latest summary from the database and displaying it on the user terminal.

[0098] The system of the present invention collects, analyzes, summarizes, and provides investment-related news quickly and efficiently. When this system is implemented, the server, terminals, and users work together to perform specific processes. A specific embodiment of this system is described below.

[0099] Gathering news information

[0100] The server configures the URL and API endpoint of a specific news source. News sources include major financial news sites and data providers. For example, the server accesses the API endpoint of a financial news site every day at 8:00 AM to retrieve the latest news data. Specifically, consider the server sending a GET request to the API endpoint "https: / / example-news-api.com / v2 / latest" to retrieve the news data in JSON format.

[0101] Preprocessing of news information

[0102] The news data acquired by the server is preprocessed before analysis. This preprocessing involves extracting only the news text and removing HTML tags and certain unnecessary characters. This converts the data into a format suitable for natural language processing.

[0103] News information analysis

[0104] The server analyzes the preprocessed news data using natural language processing (NLP) technology. SpaCy or NLTK can be used as an NLP library. NLP technology analyzes the content of news articles and identifies important keywords and sentence structures. For example, the server analyzes the news article "Nikkei Stock Average Rise 2%" and extracts keywords such as "Nikkei Stock Average," "Stock Price," and "2% Rise."

[0105] Extraction of important information

[0106] Based on the analysis results, the server filters and extracts important information necessary for investment decisions. Important information includes the causes of stock price fluctuations and corporate earnings announcements. For example, the server extracts specific information such as "The main reason for the 2% rise in the Nikkei average is the improved performance of export companies."

[0107] Generate a summary

[0108] Based on the key information extracted by the server, a summarization algorithm is used to generate summaries that are easy for investors to understand. Abstractive and extractive summarization techniques are used to generate summaries. For example, the server might generate a summary such as, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0109] Save Summary

[0110] The server saves the generated summary in a database. The data saved in the database includes the summary content, a link to the original news article, the date and time the summary was generated, etc. Specifically, the server saves the summary "The Nikkei average rose 2%, mainly due to improved performance of export companies" in an SQL database.

[0111] Provision to users

[0112] When a user accesses a financial portal site, the server retrieves the latest summary from the database and provides it to the user's device. The user can make investment decisions based on this summary. For example, when a user accesses the "Latest News" section of a financial portal site, a summary provided by the server is displayed: "The Nikkei average rose 2%, mainly due to improved performance of export companies."

[0113] Example prompts for generative AI models

[0114] By inputting the following prompt sentences into the generative AI model, the program's processing can be made easier to understand.

[0115] Example: "Describe a program that retrieves the latest investment-related news from financial news sites, uses natural language processing techniques to extract key points, and generates summaries that are easy for investors to understand."

[0116] This will enable investors to obtain the necessary information quickly and efficiently and make investment decisions.

[0117] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0118] Step 1:

[0119] The server loads the news source configuration. The server reads a previously saved configuration file and obtains the URLs and API endpoints of multiple news sources. The input is the contents of the configuration file, and the output is a list of news sources. Specifically, the server parses the "config / news_sources.json" file and stores the endpoints of each news source in a list.

[0120] Step 2:

[0121] The server periodically collects news data. Using the endpoints of the loaded news sources, the server sends API requests daily or at regular intervals to obtain the latest news data. The input is a list of news sources, and the output is news data in JSON format. Specifically, the server sends a GET request to each API endpoint at 8:00 every day and saves the received data as JSON.

[0122] Step 3:

[0123] The server preprocesses the acquired news data. The server removes unnecessary information from the acquired news data and cleans up the text data. The input is news data in JSON format, and the output is text data suitable for analysis. Specifically, the server parses the JSON data to extract the news text and removes HTML tags and unnecessary characters.

[0124] Step 4:

[0125] The server analyzes the preprocessed news data using natural language processing technology. The server uses an NLP library (for example, spaCy or NLTK) to analyze the content of the news article and identify important keywords and sentence structure. The input is the preprocessed text data, and the output is the analysis results (keywords and sentence structure). Specifically, the server analyzes the news article "Nikkei Stock Average Rise 2%" and extracts keywords such as "Nikkei Stock Average," "Stock Price," and "2% Rise."

[0126] Step 5:

[0127] The server extracts important information from the analysis results. Based on the results of natural language processing, the server filters and extracts important information necessary for investment decisions. The input is the analysis results, and the output is important information. Specifically, the server extracts information such as "The main reason for the 2% rise in the Nikkei average is the improved performance of export companies."

[0128] Step 6:

[0129] The server generates a summary based on the important information. Using a summarization algorithm, the server extracts the important information and creates a summary in a format that is easy for investors to understand. The input is the important information, and the output is the generated summary. Specifically, the server generates the summary, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0130] Step 7:

[0131] The server saves the generated summary in a database. The server stores the generated summary and its related information in a database. The input is the generated summary and its related information, and the output is the result saved in the database. Specifically, the server saves the summary "Nikkei average rises 2%, mainly due to improved performance of export companies" in an SQL database.

[0132] Step 8:

[0133] When a user accesses the portal site, the server provides the latest summary to the user's device. The server retrieves the latest summary from the database and sends it to the user's device. The input is the user's request, and the output is the summary data. In concrete terms, when a user accesses the "Latest News" section of the portal site, the server displays the summary "Nikkei average rises 2%, mainly due to improved performance of export companies."

[0134] (Application example 1)

[0135] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0136] There is a need to quickly and efficiently collect, analyze, summarize, and provide investors with a vast amount of investment-related news information. However, performing these tasks manually takes time and effort, and there is a risk of missing important timing for investment decisions. In addition, technology is needed to unify information management from different news sources and appropriately extract and summarize important information.

[0137] 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.

[0138] In this invention, the server includes means for periodically collecting news information, means for analyzing the collected news information using natural language processing, and means for extracting important data from the analyzed information, thereby enabling investment-related news to be collected, analyzed, and summarized quickly and efficiently, and provided to investors in a timely manner.

[0139] "News information" refers to current events and information collected from specific news sources.

[0140] "Periodic collection means" refers to a function that automatically obtains news information at specific time intervals.

[0141] "Means of analysis using natural language processing" refers to a technology that converts text data into a form that a computer can understand and analyzes the content.

[0142] "Means for extracting important data" refers to the function of identifying and extracting important information necessary for investment decisions from analyzed news information.

[0143] "Means for generating a summary" refers to a function that concisely summarizes the content based on the extracted important information.

[0144] "Means for storing in a database" refers to the function of storing the generated summaries in an organized manner so that they can be searched and referenced later.

[0145] "Means for providing to user terminal" refers to the function of displaying and transmitting the stored summary information to the user's device.

[0146] "Means for users to access summaries via a web application" means the ability for users to view news summaries on a web platform via the Internet.

[0147] A system for implementing the present invention includes multiple means for quickly and efficiently collecting, analyzing, summarizing, storing, and presenting investment-related news.

[0148] Gathering news information

[0149] The server pre-configures the API endpoint of a specific news source and periodically collects news data. For example, this process can be performed using the Python Requests library. For example, the server periodically accesses the news API to retrieve the latest news data. The collected data is stored in JSON format.

[0150] News information analysis

[0151] The server analyzes the collected news data using natural language processing (NLP) techniques, such as the spaCy library in Python, to identify important keywords and sentence structures from the content of news articles.

[0152] Extraction of important information

[0153] The server then uses the analysis results to extract key data needed to make investment decisions. This is done through keyword matching and rule-based filtering. For example, keywords such as "stock price" and "rise" are extracted from news articles.

[0154] Generate a summary

[0155] The server uses a summarization algorithm to generate a summary based on the extracted important information. This process uses the Python transformers library (e.g., the BERT model). For example, a summary such as "Stock prices rose 2% due to improved performance of export companies" is generated.

[0156] Save Summary

[0157] The server stores the generated summary in an SQL database. The stored data includes the summary content, a link to the original news article, the date and time the summary was generated, etc. For example, a summary such as "Stock prices rose 2% due to improved performance of export companies" is stored in the database.

[0158] Providing a summary

[0159] Users can access news summaries through a web application, which uses the Python Flask framework. For example, when a user accesses the web application, the server retrieves the latest summaries from the database and provides them to the user.

[0160] Hardware and software used

[0161] Hardware: Servers, database servers, web servers

[0162] Software: Python, Flask, spaCy, transformers, sqlite3

[0163] Prompt Sentence Examples

[0164] "Implement a system to collect the latest investment news, analyze it, extract key information, and generate summaries to provide to users."

[0165] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0166] Step 1:

[0167] The server accesses the news source's API endpoint and periodically retrieves news information. The input is the news source's API endpoint, and the output is news data in JSON format. The server uses Python's requests library to send requests to the news API and retrieves the news data received as a response.

[0168] Step 2:

[0169] The server analyzes the acquired news data using natural language processing (NLP) techniques. The input is news data in JSON format, and the output is a list of important keywords contained in the news articles. The server uses the Python spaCy library to extract important keywords and sentence structure from the content of the news articles. Specifically, it analyzes the text of the news articles and performs tokenization and part-of-speech tagging.

[0170] Step 3:

[0171] The server extracts key data based on the analysis results. The input is the analysis results of the news article, and the output is a list of key information needed to make investment decisions. Through keyword matching and rule-based filtering, the server identifies and extracts information containing key keywords such as "stock price," "rise," and "performance."

[0172] Step 4:

[0173] The server uses a summarization algorithm to generate a summary based on the extracted key information. The input is a list of key information, and the output is the summarized text. The server uses the Python transformers library and a generative AI model (e.g., the BERT model) to generate a concise summary of the key information.

[0174] Step 5:

[0175] The server stores the generated summaries in an SQL database. The input is the summarized text, and the output is the records stored in the database. The server uses Python's sqlite3 library to store the summary content, a link to the original news article, and the date and time the summary was generated in the database.

[0176] Step 6:

[0177] A user accesses the latest news summaries through a web application. The input is an access request from the user, and the output is a user interface that displays the latest news summaries. The server uses the Python Flask framework to retrieve the latest summaries from the database and provide them to the user's device. Specifically, when a user sends a request through the web application, the server queries the database, retrieves the latest summaries, and displays them to the user.

[0178] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0179] The system of the present invention not only quickly and efficiently collects, analyzes, summarizes, and provides investment-related news, but also recognizes user emotions and adaptively provides information based on those emotions. To implement this system, the following specific forms are considered and implemented.

[0180] Gathering news information

[0181] News gathering methods

[0182] The server pre-configures the URLs and API endpoints of specific news sources (e.g., major financial news sites or data providers) and periodically collects news data based on these. For example, the server retrieves the latest news from the Wall Street Journal or Reuters APIs at 8:00 every day.

[0183] Example: The server accesses the API endpoint "https: / / example-news-api.com / v2 / latest" to retrieve the latest news data in JSON format.

[0184] News information analysis

[0185] Natural language processing tools

[0186] The server analyzes the collected news data using natural language processing (NLP) techniques. NLP libraries (e.g., spaCy and NLTK) are used to analyze the content of news articles and identify important keywords and sentence structures.

[0187] Example: The server analyzes the news article "Nikkei average stock price rises 2%" and extracts keywords such as "Nikkei average," "stock price," and "2% rise."

[0188] Extraction of important information

[0189] Important information extraction means

[0190] Based on the analysis results, the server filters and extracts important information necessary for investment decisions (for example, causes of stock price fluctuations, corporate earnings announcements, etc.).

[0191] Example: From the news data analyzed by the server, specific information such as "The main reason for the 2% rise in the Nikkei average stock price is the improved performance of export companies" is extracted.

[0192] Generate a summary

[0193] Summary generation means

[0194] Based on the key information extracted by the server, a summarization algorithm is used to generate a summary that is easy for investors to understand. Abstractive summarization and extractive summarization techniques can be combined.

[0195] Example: The server generates a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0196] Save Summary

[0197] Database storage means

[0198] The server stores the generated summaries in a database, including the summary content, a link to the original news article, and the date and time the summary was generated.

[0199] Example: The server stores in an SQL database the summary "The Nikkei average rose 2%, primarily due to improved performance of export companies" along with further information.

[0200] Provision to users

[0201] User terminal provision means

[0202] When a user accesses the Yahoo Finance portal, the server retrieves the latest summary from the database and provides it to the user's device, allowing the user to make investment decisions based on it.

[0203] Example: When a user visits the "Latest News" section of Yahoo Finance, they are presented with a server-provided summary that reads, "Nikkei average rises 2%, primarily due to improved performance of export companies."

[0204] Introducing the Emotion Engine

[0205] Emotion Engine Means

[0206] An emotion engine is installed on the server to recognize how the user feels about the provided summary information. The emotion engine analyzes the user's operation history, browsing history, click behavior, browsing time, etc. to evaluate the user's emotional state.

[0207] Example: The emotion engine analyzes the user's past behavior history and determines that they tend to have positive emotions toward the news that "the Nikkei average has risen 2%."

[0208] Adaptive Delivery Methods

[0209] Based on the user's emotional state analyzed by the emotion engine, the server adjusts the content and presentation of the summary information provided to the user, for example, by prioritizing the display of content that the user feels positive about.

[0210] Example: Prioritize displaying news about "improved performance of export companies," which users have positive feelings about, along with other related news.

[0211] The system of the present invention includes these means and quickly and efficiently collects, analyzes, summarizes, and provides investment-related news, and also adaptively provides information taking into account the user's emotional state, thereby realizing more personalized investment information provision.

[0212] The processing flow will be explained below.

[0213] Specific processing steps of the news gathering and analysis system (including the sentiment engine)

[0214] Step 1:

[0215] The server configures the URL and API endpoint of the news source, for example, registering the endpoint of a particular financial news site or data provider.

[0216] Step 2:

[0217] The server retrieves news data from multiple news sources according to a pre-defined schedule, for example, calling the API every day at 8:00 to retrieve the latest news.

[0218] Step 3:

[0219] The news data retrieved by the server is preprocessed, specifically by extracting text from HTML content and converting it into a parseable format.

[0220] Step 4:

[0221] The server uses natural language processing (NLP) techniques to analyze the preprocessed news data, using NLP libraries to analyze sentence structure and extract keywords.

[0222] Step 5:

[0223] The server then uses the analysis results to extract key information from news articles, filtering the information based on specific keywords such as "stock prices," "earnings announcements," and "market trends."

[0224] Step 6:

[0225] The server organizes the extracted key information and applies a summary generation algorithm to generate summaries that are easy for investors to understand. Abstractive summaries and extractive summaries are used.

[0226] Step 7:

[0227] The server stores the generated summary in a database, including the summary content, the URL of the original article, and the date and time the summary was generated.

[0228] Step 8:

[0229] A user visits the Yahoo! Finance portal and requests to view breaking news.

[0230] Step 9:

[0231] The server retrieves the latest summary data from the database and sends it to the user terminal.

[0232] Step 10:

[0233] The user's device displays the received summary data on the Yahoo Finance interface, allowing the user to view the information.

[0234] Specific steps for implementing the Emotion Engine

[0235] Step 11:

[0236] The server collects the user's operation history, browsing history, click behavior, browsing time, etc. using an emotion engine.

[0237] Step 12:

[0238] The server analyzes the collected user data and evaluates their emotional state, for example classifying them as positive, negative, or neutral.

[0239] Step 13:

[0240] The server adjusts the content and presentation of the summary information provided based on the user's emotional state, giving priority to content that the user has a positive feeling about.

[0241] Step 14:

[0242] The user views the summary information provided preferentially and accumulates their operation history. The server then analyzes this data using the emotion engine and reflects it in the next information provided.

[0243] Specific examples

[0244] Examples:

[0245] The server retrieves news data such as "Nikkei Stock Average rises 2%" from the financial news API. The server performs NLP analysis to extract keywords such as "Nikkei Stock Average," "stock price," and "2% rise," and applies a summary generation algorithm to generate a summary such as "Nikkei Stock Average rises 2%, primarily due to improved performance of export companies." The server stores this summary in a database, and the user accesses Yahoo! Finance. The server then analyzes the user's past operation history using an emotion engine, and prioritizes displaying news with positive sentiment. This provides users with information they need to make immediate investment decisions.

[0246] Example 2

[0247] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0248] Conventional news information gathering and analysis systems collect, analyze, and summarize news, but they have the problem of being unable to provide information that takes into account the user's emotional state. As a result, it is difficult to provide the information that the user actually needs at the appropriate time, and this can result in excessive or insufficient provision of information.

[0249] 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 means for periodically collecting news data, means for analyzing the collected news data using natural language processing technology, means for extracting important information from the analyzed data, means for generating a summary based on the extracted information, means for saving the generated summary in a storage device, means for providing the saved summary to a user terminal, means for recognizing and evaluating the emotional state of a user, and means for adaptively providing information based on the evaluation result. This makes it possible to provide information adaptively taking the emotional state of a user into consideration.

[0250] "News data" is text data about current events and information, obtained from multiple news sources.

[0251] "Natural language processing technology" is a technology for processing human language using a computer, and involves analyzing text and extracting keywords.

[0252] "Storage" refers to a storage device or database for storing generated summaries.

[0253] A "usage terminal" is a device such as a computer, smartphone, or tablet that a user uses to receive information.

[0254] "User's emotional state" refers to the positive or negative emotional state that a user feels when viewing news information.

[0255] "Adaptive information provision" refers to providing information in a form optimized for a user's emotional state, taking into account that user's emotional state.

[0256] "News source" refers to the source of information that provides news data, including online newspapers, news sites, API services, etc.

[0257] This invention is a system that not only quickly and efficiently collects, analyzes, summarizes, and provides investment-related news, but also recognizes the user's emotional state and adaptively provides information based on that.

[0258] The server periodically collects news data using pre-configured news source URLs or API endpoints. For example, it uses the API endpoints of major financial news sites and data providers. Specifically, the server accesses the Wall Street Journal or Reuters API endpoint "https: / / example-news-api.com / v2 / latest" at 8:00 every day, and retrieves the latest news in JSON format.

[0259] Next, the server analyzes the collected news data using natural language processing (NLP) technology. NLP libraries such as spaCy and NLTK are used to analyze the content of news articles and identify important keywords and sentence structures. For example, the server analyzes the news article "Nikkei Stock Average Rise 2%" and extracts the keywords "Nikkei Stock Average," "Stock Price," and "Rise 2%."

[0260] Next, the server extracts important information based on the analysis results. It filters and extracts information necessary for investment decisions (e.g., causes of stock price fluctuations, corporate earnings announcements, etc.). For example, the server might extract specific information such as, "The main reason the Nikkei average rose 2% is the improved performance of export companies."

[0261] The server then uses a summarization algorithm based on the key information extracted to generate a summary that is easy for investors to understand. Abstractive and extractive summarization techniques can be combined. For example, the server generates a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0262] The server stores the generated summary in a storage device, such as an SQL database. The stored data includes the summary content, a link to the original news article, the date and time the summary was generated, etc. For example, the server stores the summary "The Nikkei average rose 2%, mainly due to improved performance of export companies" and its detailed information in the SQL database.

[0263] When a user accesses the Yahoo! Finance portal, the server retrieves the latest summary from the database and provides it to the user's device. The user can then make investment decisions based on this summary. For example, when a user accesses the "Latest News" section of Yahoo! Finance, the server provides a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0264] Furthermore, the server is equipped with an emotion engine that recognizes and evaluates the user's emotional state. The emotion engine analyzes the user's operation history, browsing history, click behavior, browsing time, etc. to evaluate the user's emotional state. For example, the emotion engine analyzes the user's past operation history and determines that the user tends to have positive emotions toward the news that "the Nikkei average rose 2%."

[0265] Based on the user's emotional state analyzed by the emotion engine, the server adjusts the content and display method of the summary information provided to the user's device. It sets the display to prioritize content that the user has a positive feeling about. For example, news about "improving performance of export companies," which the user has a positive feeling about, is displayed with priority along with other related news.

[0266] Prompt Sentence Examples

[0267] "Collect the latest financial news from the Wall Street Journal and extract and summarize the key information."

[0268] "Analyze this news article and identify key keywords."

[0269] "If a user has a positive sentiment towards a particular news item, prioritize showing other related news items."

[0270] In this way, the system of the present invention performs all processes from collecting news data to providing information to the user in one place, and realizes adaptive information provision based on the emotional state of the user.

[0271] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0272] Step 1: Gathering news data

[0273] The server loads the URL or API endpoint of a pre-configured news source, sends a request to the news source's API at the specified time, and retrieves the latest news. Specifically, the server sends a GET request to the API endpoint (e.g., "https: / / example-news-api.com / v2 / latest") at 8:00 every day, and receives the latest news data in JSON format.

[0274] Input: News source URL and API endpoint

[0275] Output: News data in JSON format

[0276] Step 2: Analyzing the news data

[0277] The server analyzes the collected news data using natural language processing technology. An NLP library (e.g., spaCy or NLTK) is used to analyze the content of news articles and identify sentence structure and important keywords. For example, the server analyzes a news article titled "Nikkei Stock Average Rise 2%" and extracts the keywords "Nikkei Stock Average," "Stock Price," and "2% Rise."

[0278] Input: News data in JSON format

[0279] Output: Parsed text data and keywords

[0280] Step 3: Extracting important information

[0281] Based on the analysis results, the server filters and extracts important information necessary for investment decisions. Specifically, the server extracts the specific information that "the main reason for the 2% rise in the Nikkei average is the improved performance of export companies."

[0282] Input: Parsed text data and keywords

[0283] Output: Extracted important information

[0284] Step 4: Generate a summary

[0285] The server uses a summarization algorithm to generate an easy-to-understand summary based on the key information extracted. For example, the server may generate a summary such as, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0286] Input: Extracted sensitive information

[0287] Output: Generated summary

[0288] Step 5: Save the summary

[0289] The server saves the generated summary in a database. Specifically, the database stores the summary content, a link to the original news article, the date and time of generation, etc. For example, the server saves the summary "Nikkei average rises 2%, mainly due to improved performance of export companies" along with the link and the date and time of generation in an SQL database.

[0290] Input: Generated summary

[0291] Output: Summary and detailed information stored in a database

[0292] Step 6: Provide to users

[0293] When a user accesses a specified portal, the server retrieves the latest summary from the database and provides it to the user's device. For example, when a user accesses the "Latest News" section, the server displays a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0294] Input: User request and summary stored in database

[0295] Output: Latest summary displayed on user terminal

[0296] Step 7: Assess your emotional state

[0297] The server uses an emotion engine to analyze the user's operation history, browsing history, click behavior, browsing time, etc., and evaluates the user's emotional state. Specifically, the emotion engine analyzes the user's past operation data and determines that the user tends to have positive emotions toward the news that "the Nikkei average rose 2%."

[0298] Input: User operation history, browsing history, click behavior, browsing time, etc.

[0299] Output: Evaluated user emotional state

[0300] Step 8: Adaptive information provision

[0301] Based on the evaluation results of the emotion engine, the server adjusts the content and display method of the summary information provided to the user's device. For example, it prioritizes the display of news about "improving performance of export companies," which the user has positive feelings about.

[0302] Input: Evaluated emotional state of the user and summaries stored in a database

[0303] Output: Related news displayed in priority order according to the user's emotional state

[0304] (Application example 2)

[0305] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0306] Conventional investment news delivery systems are limited to analyzing collected news information and generating summaries, and do not provide information based on the user's individual emotions or interests. This results in users being exposed to a large amount of unnecessary information, making it difficult to quickly obtain the information necessary for investment decisions. Furthermore, the lack of a mechanism for adaptively providing information based on the user's emotional state makes it difficult to provide personalized investment information.

[0307] The specification processing by the specification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for periodically collecting news information, means for analyzing the collected news information using natural language processing, means for extracting important data from the analyzed information, means for generating a summary based on the extracted information, means for saving the generated summary in a database, means for providing the saved summary to a user terminal, means for recognizing the user's emotional state, and means for adaptively adjusting the information to be provided based on the user's emotional state. This enables users to quickly and efficiently obtain personalized investment information based on their own emotions and interests.

[0308] "Means for regularly collecting news information" refers to a system for obtaining news data from specific news sources at regular intervals.

[0309] "Means for analyzing collected news information using natural language processing" refers to a system that analyzes collected news data using natural language processing technology to understand the content and extract important keywords.

[0310] "Means for extracting important data from analyzed information" refers to a system that filters and extracts information and keywords that are important for investment decisions from news data analyzed using natural language processing.

[0311] A "means for generating a summary based on extracted information" is an algorithm or program that uses key data to summarize information in a way that is easy for users to understand.

[0312] The "means for storing the generated summary in a database" refers to a mechanism for storing the generated summary information in a database so that it can be accessed and searched later.

[0313] The "means for providing the stored summary to the user terminal" is a mechanism for presenting the summary information stored in the database to the user terminal.

[0314] The "means for recognizing the user's emotional state" is a mechanism that analyzes the user's operation history, browsing behavior, etc. to evaluate their current emotions and interests.

[0315] The "means for adaptively adjusting the information provided based on the emotional state" is a mechanism that dynamically changes the information provided and the way it is displayed based on the user's emotional evaluation results.

[0316] The system of the present invention quickly and efficiently collects, analyzes, and summarizes investment-related news, and adaptively provides information based on the user's sentiment. Specific means and processes for implementing this system are described below.

[0317] Gathering news information

[0318] News gathering methods

[0319] The server periodically collects news data from specific news sources, such as major financial news sites and data providers, using HTTP client software such as the requests library in Python.

[0320] News information analysis

[0321] Natural language processing tools

[0322] The server analyzes the collected news data using natural language processing (NLP) technology. NLP uses libraries such as spaCy and NLTK to analyze the content of news articles and identify important keywords and sentence structures. This process converts the content of the news data into an understandable format.

[0323] Extraction of important information

[0324] Important information extraction means

[0325] From the analyzed news data, particularly important information is extracted. This can be done using machine learning algorithms or rule-based analysis methods. For example, the causes of stock price fluctuations and corporate earnings announcements that are important for investment decisions are targeted.

[0326] Generate a summary

[0327] Summary generation means

[0328] The server generates a summary based on the extracted important information using a summary generation algorithm. The algorithm can combine abstractive and extractive summarization techniques. Generative AI models such as GPT-3 (registered trademark) are particularly effective.

[0329] Save Summary

[0330] Database storage means

[0331] The generated summaries are then stored in a database, which also contains metadata such as the summary content, a link to the original news article, and the date and time the summary was generated. This process is often performed using a cloud database service such as Amazon RDS.

[0332] Provision to users

[0333] User terminal provision means

[0334] When a user accesses the system through a device (such as a smartphone or PC), the server retrieves the latest summary from the database and provides it to the user, allowing the user to obtain important investment information in a timely manner.

[0335] Introducing the Emotion Engine

[0336] Emotion Engine Means

[0337] An emotion engine is installed on the server to evaluate how users feel about the summary information provided. The emotion engine collects and analyzes data such as user operation history, browsing history, click behavior, and browsing time. Machine learning libraries such as TENSORFLOW (registered trademark) and Keras are used for emotion analysis.

[0338] Adaptive Delivery Methods

[0339] Based on the user's emotional state analyzed by the emotion engine, the server adjusts the content and presentation of the summary information it provides. For example, it may prioritize news stories to which the user has responded positively. This dynamic adjustment is an important feature for providing information optimized for each individual user.

[0340] Specific examples

[0341] 1. News gathering: Access the API endpoint "https: / / example-news-api.com / v2 / latest" to get the latest news data.

[0342] 2. News analysis: Analyze the collected news item "Nikkei average stock price rises 2%" and extract keywords such as "Nikkei average," "stock price," and "2% rise."

[0343] 3. Extraction of important information: Extract important information such as "The main reason for the 2% rise in the Nikkei average stock price is the improved performance of export companies."

[0344] 4. Summary generation: Using GPT-3, generate a summary such as "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0345] 5. Emotion engine: Analyzes the user's past operation history and determines that they tend to have positive emotions toward the news that "Nikkei average rises 2%," and prioritizes displaying content that evokes positive emotions.

[0346] Prompt Sentence Examples

[0347] Prompt for news summary generation:

[0348] News article: "The Nikkei Stock Average rose 2%. This was mainly due to improved performance of export companies."

[0349] Generated summary: "Nikkei average rises 2%, primarily due to improved performance of export companies"

[0350] In this way, the invention can be specifically implemented.

[0351] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0352] Step 1: Get news information

[0353] The server periodically accesses the news API endpoint to retrieve the latest news data. The input is the API endpoint URL, and the output is the retrieved news data (in JSON format). Specifically, it sends an HTTP request using the Python requests library and receives the news data as a response.

[0354] Step 2: Analyzing the news information

[0355] The server analyzes the news data acquired in step 1 using natural language processing (NLP) technology. The input is the acquired news data, and the output is the analyzed important keywords and phrases. Specifically, it uses NLP libraries such as spaCy and NLTK to perform text analysis.

[0356] Step 3: Extracting important information

[0357] The server extracts important information necessary for investment decisions from the data analyzed in step 2. The input is the analyzed keywords and phrases, and the output is the extracted important information. Specifically, it uses machine learning algorithms and rule-based filtering methods to select important information.

[0358] Step 4: Generate a summary

[0359] The server generates a summary using a summary generation algorithm based on the information extracted in step 3. The input is the extracted key information, and the output is the generated summary. Specifically, a generative AI model such as GPT-3 is used to create the summary. An example of a prompt sentence for the generative AI model is, "News article: The Nikkei average stock price rose 2%. This is mainly due to the improved performance of export companies. Generated summary: The Nikkei average rose 2%, mainly due to the improved performance of export companies."

[0360] Step 5: Save the summary

[0361] The server stores the summaries generated in step 4 in a database. The input is the generated summary information, and the output is the summary data stored in the database. Specifically, the server uses an SQL query to insert the summary information into a cloud database such as Amazon RDS.

[0362] Step 6: Deliver to users

[0363] When a user accesses the system from a terminal, the server retrieves the latest summary from the database and provides it to the user's terminal. The input is the user's access request, and the output is the displayed summary information. Specifically, it receives the user's request, retrieves the summary information from the database using an SQL query, and returns it as an HTTP response.

[0364] Step 7: Recognizing your emotional state

[0365] The server analyzes the user's operation history and browsing behavior to evaluate their emotional state. The input is the user's past operation data, and the output is the evaluation result of the user's emotional state. Specifically, emotion analysis is performed using machine learning libraries such as TensorFlow and Keras.

[0366] Step 8: Adaptively adjusting information provision

[0367] The server adjusts the content and presentation of the information it provides based on the emotion evaluation results obtained in step 7. The input is the evaluation result of the emotional state, and the output is the adjusted summary information. Specifically, it reorganizes the list of high-priority information and applies an algorithm to display it in a form optimized for the user.

[0368] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0369] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0370] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0371] [Second embodiment]

[0372] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0373] 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.

[0374] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0375] 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.

[0376] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0377] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0378] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0379] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0380] 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.

[0381] 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.

[0382] 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.

[0383] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0384] The system of the present invention collects, analyzes, summarizes, and provides investment-related news quickly and efficiently. To implement this system, the following specific forms are considered and implemented.

[0385] Gathering news information

[0386] News gathering methods

[0387] The server pre-configures the URLs and API endpoints of specific news sources (e.g., major financial news sites or data providers) and periodically collects news data based on these. For example, the server retrieves the latest news from the Wall Street Journal or Reuters APIs at 8:00 every day.

[0388] Example: The server accesses the API endpoint "https: / / example-news-api.com / v2 / latest" to retrieve the latest news data in JSON format.

[0389] News information analysis

[0390] Natural language processing tools

[0391] The server analyzes the collected news data using natural language processing (NLP) techniques. NLP libraries (e.g., spaCy and NLTK) are used to analyze the content of news articles and identify important keywords and sentence structures.

[0392] Example: The server analyzes the news article "Nikkei average stock price rises 2%" and extracts keywords such as "Nikkei average," "stock price," and "2% rise."

[0393] Extraction of important information

[0394] Important information extraction means

[0395] Based on the analysis results, the server filters and extracts important information necessary for investment decisions (for example, causes of stock price fluctuations, corporate earnings announcements, etc.).

[0396] Example: From the news data analyzed by the server, specific information such as "The main reason for the 2% rise in the Nikkei average stock price is the improved performance of export companies" is extracted.

[0397] Generate a summary

[0398] Summary generation means

[0399] Based on the key information extracted by the server, a summarization algorithm is used to generate a summary that is easy for investors to understand. Abstractive summarization and extractive summarization techniques can be combined.

[0400] Example: The server generates a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0401] Save Summary

[0402] Database storage means

[0403] The server stores the generated summaries in a database, including the summary content, a link to the original news article, and the date and time the summary was generated.

[0404] Example: The server stores in an SQL database the summary "The Nikkei average rose 2%, primarily due to improved performance of export companies" along with further information.

[0405] Provision to users

[0406] User terminal provision means

[0407] When a user accesses the Yahoo Finance portal, the server retrieves the latest summary from the database and provides it to the user's device, allowing the user to make investment decisions based on it.

[0408] Example: When a user visits the "Latest News" section of Yahoo Finance, they are presented with a server-provided summary that reads, "Nikkei average rises 2%, primarily due to improved performance of export companies."

[0409] The system of the present invention includes these means and is capable of collecting, analyzing, summarizing and providing investment-related news quickly and efficiently, thereby greatly assisting investors in obtaining information.

[0410] The processing flow will be explained below.

[0411] Specific processing steps of the news gathering and analysis system

[0412] Step 1:

[0413] The server configures the URL and API endpoint of the news source, for example, registering the endpoint of a particular financial news site or data provider.

[0414] Step 2:

[0415] The server retrieves news data from multiple news sources according to a pre-defined schedule, for example, calling the API every day at 8:00 to retrieve the latest news.

[0416] Step 3:

[0417] The news data retrieved by the server is preprocessed, specifically by extracting text from HTML content and converting it into a parseable format.

[0418] Step 4:

[0419] The server uses natural language processing (NLP) techniques to analyze the preprocessed news data, using NLP libraries to analyze sentence structure and extract keywords.

[0420] Step 5:

[0421] The server then uses the analysis results to extract key information from news articles, filtering the information based on specific keywords such as "stock prices," "earnings announcements," and "market trends."

[0422] Step 6:

[0423] The server organizes the extracted key information and applies a summary generation algorithm to generate summaries that are easy for investors to understand. Abstractive summaries and extractive summaries are used.

[0424] Step 7:

[0425] The server stores the generated summary in a database, including the summary content, the URL of the original article, and the date and time the summary was generated.

[0426] Step 8:

[0427] A user visits the Yahoo! Finance portal and requests to view breaking news.

[0428] Step 9:

[0429] The server retrieves the latest summary data from the database and sends it to the user terminal.

[0430] Step 10:

[0431] The user's device displays the received summary data on the Yahoo Finance interface, allowing the user to view the information.

[0432] Example 1

[0433] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0434] Because news information is vast and scattered, investors do not have time to research and analyze it individually, making it difficult to quickly and efficiently obtain important news information. Furthermore, analyzing news information and generating summaries requires advanced technology and knowledge, and doing this manually is labor-intensive. Therefore, there is a need for a system that can quickly and accurately provide the important news information investors need.

[0435] 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.

[0436] In this invention, the server includes means for periodically collecting news information, means for preprocessing the collected news information, means for analyzing the preprocessed news information using natural language processing, means for extracting important data from the analyzed information, means for generating summaries based on the extracted important data, means for storing the generated summaries in a database, and means for providing the stored summaries to a user terminal, thereby enabling investors to quickly and efficiently obtain the latest important news information and make investment decisions.

[0437] "Means for regularly collecting news information" refers to a function that allows you to set the URL or API endpoint of a specific news source and obtain news data daily or at a fixed time.

[0438] "Means for preprocessing collected news information" refers to a function for removing unnecessary information from acquired news data and converting it into a format that can be analyzed.

[0439] "Means for analyzing preprocessed news information using natural language processing" refers to a function that uses natural language processing techniques and libraries to analyze preprocessed news text and identify important keywords and sentence structures.

[0440] "Means of extracting important data from analyzed information" refers to a function that filters and selects important information necessary for investment decisions from the data obtained as a result of natural language processing.

[0441] The "means for generating a summary based on the extracted important data" is a function that uses an algorithm to create a summary based on the extracted important information in a form that is easy for investors to understand.

[0442] "Means for saving the generated summary in a database" refers to a function that saves the generated summary and its related information in a database so that it can be referenced or searched later.

[0443] The "means for providing the stored summary to the user terminal" is a function for retrieving the latest summary from the database and displaying it on the user terminal.

[0444] The system of the present invention collects, analyzes, summarizes, and provides investment-related news quickly and efficiently. When this system is implemented, the server, terminals, and users work together to perform specific processes. A specific embodiment of this system is described below.

[0445] Gathering news information

[0446] The server configures the URL and API endpoint of a specific news source. News sources include major financial news sites and data providers. For example, the server accesses the API endpoint of a financial news site every day at 8:00 AM to retrieve the latest news data. Specifically, consider the server sending a GET request to the API endpoint "https: / / example-news-api.com / v2 / latest" to retrieve the news data in JSON format.

[0447] Preprocessing of news information

[0448] The news data acquired by the server is preprocessed before analysis. This preprocessing involves extracting only the news text and removing HTML tags and certain unnecessary characters. This converts the data into a format suitable for natural language processing.

[0449] News information analysis

[0450] The server analyzes the preprocessed news data using natural language processing (NLP) technology. SpaCy or NLTK can be used as an NLP library. NLP technology analyzes the content of news articles and identifies important keywords and sentence structures. For example, the server analyzes the news article "Nikkei Stock Average Rise 2%" and extracts keywords such as "Nikkei Stock Average," "Stock Price," and "2% Rise."

[0451] Extraction of important information

[0452] Based on the analysis results, the server filters and extracts important information necessary for investment decisions. Important information includes the causes of stock price fluctuations and corporate earnings announcements. For example, the server extracts specific information such as "The main reason for the 2% rise in the Nikkei average is the improved performance of export companies."

[0453] Generate a summary

[0454] Based on the key information extracted by the server, a summarization algorithm is used to generate summaries that are easy for investors to understand. Abstractive and extractive summarization techniques are used to generate summaries. For example, the server might generate a summary such as, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0455] Save Summary

[0456] The server saves the generated summary in a database. The data saved in the database includes the summary content, a link to the original news article, the date and time the summary was generated, etc. Specifically, the server saves the summary "The Nikkei average rose 2%, mainly due to improved performance of export companies" in an SQL database.

[0457] Provision to users

[0458] When a user accesses a financial portal site, the server retrieves the latest summary from the database and provides it to the user's device. The user can make investment decisions based on this summary. For example, when a user accesses the "Latest News" section of a financial portal site, a summary provided by the server is displayed: "The Nikkei average rose 2%, mainly due to improved performance of export companies."

[0459] Example prompts for generative AI models

[0460] By inputting the following prompt sentences into the generative AI model, the program's processing can be made easier to understand.

[0461] Example: "Describe a program that retrieves the latest investment-related news from financial news sites, uses natural language processing techniques to extract key points, and generates summaries that are easy for investors to understand."

[0462] This will enable investors to obtain the necessary information quickly and efficiently and make investment decisions.

[0463] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0464] Step 1:

[0465] The server loads the news source configuration. The server reads a previously saved configuration file and obtains the URLs and API endpoints of multiple news sources. The input is the contents of the configuration file, and the output is a list of news sources. Specifically, the server parses the "config / news_sources.json" file and stores the endpoints of each news source in a list.

[0466] Step 2:

[0467] The server periodically collects news data. Using the endpoints of the loaded news sources, the server sends API requests daily or at regular intervals to obtain the latest news data. The input is a list of news sources, and the output is news data in JSON format. Specifically, the server sends a GET request to each API endpoint at 8:00 every day and saves the received data as JSON.

[0468] Step 3:

[0469] The server preprocesses the acquired news data. The server removes unnecessary information from the acquired news data and cleans up the text data. The input is news data in JSON format, and the output is text data suitable for analysis. Specifically, the server parses the JSON data to extract the news text and removes HTML tags and unnecessary characters.

[0470] Step 4:

[0471] The server analyzes the preprocessed news data using natural language processing technology. The server uses an NLP library (for example, spaCy or NLTK) to analyze the content of the news article and identify important keywords and sentence structure. The input is the preprocessed text data, and the output is the analysis results (keywords and sentence structure). Specifically, the server analyzes the news article "Nikkei Stock Average Rise 2%" and extracts keywords such as "Nikkei Stock Average," "Stock Price," and "2% Rise."

[0472] Step 5:

[0473] The server extracts important information from the analysis results. Based on the results of natural language processing, the server filters and extracts important information necessary for investment decisions. The input is the analysis results, and the output is important information. Specifically, the server extracts information such as "The main reason for the 2% rise in the Nikkei average is the improved performance of export companies."

[0474] Step 6:

[0475] The server generates a summary based on the important information. Using a summarization algorithm, the server extracts the important information and creates a summary in a format that is easy for investors to understand. The input is the important information, and the output is the generated summary. Specifically, the server generates the summary, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0476] Step 7:

[0477] The server saves the generated summary in a database. The server stores the generated summary and its related information in a database. The input is the generated summary and its related information, and the output is the result saved in the database. Specifically, the server saves the summary "Nikkei average rises 2%, mainly due to improved performance of export companies" in an SQL database.

[0478] Step 8:

[0479] When a user accesses the portal site, the server provides the latest summary to the user's device. The server retrieves the latest summary from the database and sends it to the user's device. The input is the user's request, and the output is the summary data. In concrete terms, when a user accesses the "Latest News" section of the portal site, the server displays the summary "Nikkei average rises 2%, mainly due to improved performance of export companies."

[0480] (Application example 1)

[0481] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0482] There is a need to quickly and efficiently collect, analyze, summarize, and provide investors with a vast amount of investment-related news information. However, performing these tasks manually takes time and effort, and there is a risk of missing important timing for investment decisions. In addition, technology is needed to unify information management from different news sources and appropriately extract and summarize important information.

[0483] 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.

[0484] In this invention, the server includes means for periodically collecting news information, means for analyzing the collected news information using natural language processing, and means for extracting important data from the analyzed information, thereby enabling investment-related news to be collected, analyzed, and summarized quickly and efficiently, and provided to investors in a timely manner.

[0485] "News information" refers to current events and information collected from specific news sources.

[0486] "Periodic collection means" refers to a function that automatically obtains news information at specific time intervals.

[0487] "Means of analysis using natural language processing" refers to a technology that converts text data into a form that a computer can understand and analyzes the content.

[0488] "Means for extracting important data" refers to the function of identifying and extracting important information necessary for investment decisions from analyzed news information.

[0489] "Means for generating a summary" refers to a function that concisely summarizes the content based on the extracted important information.

[0490] "Means for storing in a database" refers to the function of storing the generated summaries in an organized manner so that they can be searched and referenced later.

[0491] "Means for providing to user terminal" refers to the function of displaying and transmitting the stored summary information to the user's device.

[0492] "Means for users to access summaries via a web application" means the ability for users to view news summaries on a web platform via the Internet.

[0493] A system for implementing the present invention includes multiple means for quickly and efficiently collecting, analyzing, summarizing, storing, and presenting investment-related news.

[0494] Gathering news information

[0495] The server pre-configures the API endpoint of a specific news source and periodically collects news data. For example, this process can be performed using the Python Requests library. For example, the server periodically accesses the news API to retrieve the latest news data. The collected data is stored in JSON format.

[0496] News information analysis

[0497] The server analyzes the collected news data using natural language processing (NLP) techniques, such as the spaCy library in Python, to identify important keywords and sentence structures from the content of news articles.

[0498] Extraction of important information

[0499] The server then uses the analysis results to extract key data needed to make investment decisions. This is done through keyword matching and rule-based filtering. For example, keywords such as "stock price" and "rise" are extracted from news articles.

[0500] Generate a summary

[0501] The server uses a summarization algorithm to generate a summary based on the extracted important information. This process uses the Python transformers library (e.g., the BERT model). For example, a summary such as "Stock prices rose 2% due to improved performance of export companies" is generated.

[0502] Save Summary

[0503] The server stores the generated summary in an SQL database. The stored data includes the summary content, a link to the original news article, the date and time the summary was generated, etc. For example, a summary such as "Stock prices rose 2% due to improved performance of export companies" is stored in the database.

[0504] Providing a summary

[0505] Users can access news summaries through a web application, which uses the Python Flask framework. For example, when a user accesses the web application, the server retrieves the latest summaries from the database and provides them to the user.

[0506] Hardware and software used

[0507] Hardware: Servers, database servers, web servers

[0508] Software: Python, Flask, spaCy, transformers, sqlite3

[0509] Prompt Sentence Examples

[0510] "Implement a system to collect the latest investment news, analyze it, extract key information, and generate summaries to provide to users."

[0511] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0512] Step 1:

[0513] The server accesses the news source's API endpoint and periodically retrieves news information. The input is the news source's API endpoint, and the output is news data in JSON format. The server uses Python's requests library to send requests to the news API and retrieves the news data received as a response.

[0514] Step 2:

[0515] The server analyzes the acquired news data using natural language processing (NLP) techniques. The input is news data in JSON format, and the output is a list of important keywords contained in the news articles. The server uses the Python spaCy library to extract important keywords and sentence structure from the content of the news articles. Specifically, it analyzes the text of the news articles and performs tokenization and part-of-speech tagging.

[0516] Step 3:

[0517] The server extracts key data based on the analysis results. The input is the analysis results of the news article, and the output is a list of key information needed to make investment decisions. Through keyword matching and rule-based filtering, the server identifies and extracts information containing key keywords such as "stock price," "rise," and "performance."

[0518] Step 4:

[0519] The server uses a summarization algorithm to generate a summary based on the extracted key information. The input is a list of key information, and the output is the summarized text. The server uses the Python transformers library and a generative AI model (e.g., the BERT model) to generate a concise summary of the key information.

[0520] Step 5:

[0521] The server stores the generated summaries in an SQL database. The input is the summarized text, and the output is the records stored in the database. The server uses Python's sqlite3 library to store the summary content, a link to the original news article, and the date and time the summary was generated in the database.

[0522] Step 6:

[0523] A user accesses the latest news summaries through a web application. The input is an access request from the user, and the output is a user interface that displays the latest news summaries. The server uses the Python Flask framework to retrieve the latest summaries from the database and provide them to the user's device. Specifically, when a user sends a request through the web application, the server queries the database, retrieves the latest summaries, and displays them to the user.

[0524] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0525] The system of the present invention not only quickly and efficiently collects, analyzes, summarizes, and provides investment-related news, but also recognizes user emotions and adaptively provides information based on those emotions. To implement this system, the following specific forms are considered and implemented.

[0526] Gathering news information

[0527] News gathering methods

[0528] The server pre-configures the URLs and API endpoints of specific news sources (e.g., major financial news sites or data providers) and periodically collects news data based on these. For example, the server retrieves the latest news from the Wall Street Journal or Reuters APIs at 8:00 every day.

[0529] Example: The server accesses the API endpoint "https: / / example-news-api.com / v2 / latest" to retrieve the latest news data in JSON format.

[0530] News information analysis

[0531] Natural language processing tools

[0532] The server analyzes the collected news data using natural language processing (NLP) techniques. NLP libraries (e.g., spaCy and NLTK) are used to analyze the content of news articles and identify important keywords and sentence structures.

[0533] Example: The server analyzes the news article "Nikkei average stock price rises 2%" and extracts keywords such as "Nikkei average," "stock price," and "2% rise."

[0534] Extraction of important information

[0535] Important information extraction means

[0536] Based on the analysis results, the server filters and extracts important information necessary for investment decisions (for example, causes of stock price fluctuations, corporate earnings announcements, etc.).

[0537] Example: From the news data analyzed by the server, specific information such as "The main reason for the 2% rise in the Nikkei average stock price is the improved performance of export companies" is extracted.

[0538] Generate a summary

[0539] Summary generation means

[0540] Based on the key information extracted by the server, a summarization algorithm is used to generate a summary that is easy for investors to understand. Abstractive summarization and extractive summarization techniques can be combined.

[0541] Example: The server generates a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0542] Save Summary

[0543] Database storage means

[0544] The server stores the generated summaries in a database, including the summary content, a link to the original news article, and the date and time the summary was generated.

[0545] Example: The server stores in an SQL database the summary "The Nikkei average rose 2%, primarily due to improved performance of export companies" along with further information.

[0546] Provision to users

[0547] User terminal provision means

[0548] When a user accesses the Yahoo Finance portal, the server retrieves the latest summary from the database and provides it to the user's device, allowing the user to make investment decisions based on it.

[0549] Example: When a user visits the "Latest News" section of Yahoo Finance, they are presented with a server-provided summary that reads, "Nikkei average rises 2%, primarily due to improved performance of export companies."

[0550] Introducing the Emotion Engine

[0551] Emotion Engine Means

[0552] An emotion engine is installed on the server to recognize how the user feels about the provided summary information. The emotion engine analyzes the user's operation history, browsing history, click behavior, browsing time, etc. to evaluate the user's emotional state.

[0553] Example: The emotion engine analyzes the user's past behavior history and determines that they tend to have positive emotions toward the news that "the Nikkei average has risen 2%."

[0554] Adaptive Delivery Methods

[0555] Based on the user's emotional state analyzed by the emotion engine, the server adjusts the content and presentation of the summary information provided to the user, for example, by prioritizing the display of content that the user feels positive about.

[0556] Example: Prioritize displaying news about "improved performance of export companies," which users have positive feelings about, along with other related news.

[0557] The system of the present invention includes these means and quickly and efficiently collects, analyzes, summarizes, and provides investment-related news, and also adaptively provides information taking into account the user's emotional state, thereby realizing more personalized investment information provision.

[0558] The processing flow will be explained below.

[0559] Specific processing steps of the news gathering and analysis system (including the sentiment engine)

[0560] Step 1:

[0561] The server configures the URL and API endpoint of the news source, for example, registering the endpoint of a particular financial news site or data provider.

[0562] Step 2:

[0563] The server retrieves news data from multiple news sources according to a pre-defined schedule, for example, calling the API every day at 8:00 to retrieve the latest news.

[0564] Step 3:

[0565] The news data retrieved by the server is preprocessed, specifically by extracting text from HTML content and converting it into a parseable format.

[0566] Step 4:

[0567] The server uses natural language processing (NLP) techniques to analyze the preprocessed news data, using NLP libraries to analyze sentence structure and extract keywords.

[0568] Step 5:

[0569] The server then uses the analysis results to extract key information from news articles, filtering the information based on specific keywords such as "stock prices," "earnings announcements," and "market trends."

[0570] Step 6:

[0571] The server organizes the extracted key information and applies a summary generation algorithm to generate summaries that are easy for investors to understand. Abstractive summaries and extractive summaries are used.

[0572] Step 7:

[0573] The server stores the generated summary in a database, including the summary content, the URL of the original article, and the date and time the summary was generated.

[0574] Step 8:

[0575] A user visits the Yahoo! Finance portal and requests to view breaking news.

[0576] Step 9:

[0577] The server retrieves the latest summary data from the database and sends it to the user terminal.

[0578] Step 10:

[0579] The user's device displays the received summary data on the Yahoo Finance interface, allowing the user to view the information.

[0580] Specific steps for implementing the Emotion Engine

[0581] Step 11:

[0582] The server collects the user's operation history, browsing history, click behavior, browsing time, etc. using an emotion engine.

[0583] Step 12:

[0584] The server analyzes the collected user data and evaluates their emotional state, for example classifying them as positive, negative, or neutral.

[0585] Step 13:

[0586] The server adjusts the content and presentation of the summary information provided based on the user's emotional state, giving priority to content that the user has a positive feeling about.

[0587] Step 14:

[0588] The user views the summary information provided preferentially and accumulates their operation history. The server then analyzes this data using the emotion engine and reflects it in the next information provided.

[0589] Specific examples

[0590] Examples:

[0591] The server retrieves news data such as "Nikkei Stock Average rises 2%" from the financial news API. The server performs NLP analysis to extract keywords such as "Nikkei Stock Average," "stock price," and "2% rise," and applies a summary generation algorithm to generate a summary such as "Nikkei Stock Average rises 2%, primarily due to improved performance of export companies." The server stores this summary in a database, and the user accesses Yahoo! Finance. The server then analyzes the user's past operation history using an emotion engine, and prioritizes displaying news with positive sentiment. This provides users with information they need to make immediate investment decisions.

[0592] Example 2

[0593] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0594] Conventional news information gathering and analysis systems collect, analyze, and summarize news, but they have the problem of being unable to provide information that takes into account the user's emotional state. As a result, it is difficult to provide the information that the user actually needs at the appropriate time, and this can result in excessive or insufficient provision of information.

[0595] 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 means for periodically collecting news data, means for analyzing the collected news data using natural language processing technology, means for extracting important information from the analyzed data, means for generating a summary based on the extracted information, means for saving the generated summary in a storage device, means for providing the saved summary to a user terminal, means for recognizing and evaluating the emotional state of a user, and means for adaptively providing information based on the evaluation result. This makes it possible to provide information adaptively taking the emotional state of a user into consideration.

[0596] "News data" is text data about current events and information, obtained from multiple news sources.

[0597] "Natural language processing technology" is a technology for processing human language using a computer, and involves analyzing text and extracting keywords.

[0598] "Storage" refers to a storage device or database for storing generated summaries.

[0599] A "usage terminal" is a device such as a computer, smartphone, or tablet that a user uses to receive information.

[0600] "User's emotional state" refers to the positive or negative emotional state that a user feels when viewing news information.

[0601] "Adaptive information provision" refers to providing information in a form optimized for a user's emotional state, taking into account that user's emotional state.

[0602] "News source" refers to the source of information that provides news data, including online newspapers, news sites, API services, etc.

[0603] This invention is a system that not only quickly and efficiently collects, analyzes, summarizes, and provides investment-related news, but also recognizes the user's emotional state and adaptively provides information based on that.

[0604] The server periodically collects news data using pre-configured news source URLs or API endpoints. For example, it uses the API endpoints of major financial news sites and data providers. Specifically, the server accesses the Wall Street Journal or Reuters API endpoint "https: / / example-news-api.com / v2 / latest" at 8:00 every day, and retrieves the latest news in JSON format.

[0605] Next, the server analyzes the collected news data using natural language processing (NLP) technology. NLP libraries such as spaCy and NLTK are used to analyze the content of news articles and identify important keywords and sentence structures. For example, the server analyzes the news article "Nikkei Stock Average Rise 2%" and extracts the keywords "Nikkei Stock Average," "Stock Price," and "Rise 2%."

[0606] Next, the server extracts important information based on the analysis results. It filters and extracts information necessary for investment decisions (e.g., causes of stock price fluctuations, corporate earnings announcements, etc.). For example, the server might extract specific information such as, "The main reason the Nikkei average rose 2% is the improved performance of export companies."

[0607] The server then uses a summarization algorithm based on the key information extracted to generate a summary that is easy for investors to understand. Abstractive and extractive summarization techniques can be combined. For example, the server generates a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0608] The server stores the generated summary in a storage device, such as an SQL database. The stored data includes the summary content, a link to the original news article, the date and time the summary was generated, etc. For example, the server stores the summary "The Nikkei average rose 2%, mainly due to improved performance of export companies" and its detailed information in the SQL database.

[0609] When a user accesses the Yahoo! Finance portal, the server retrieves the latest summary from the database and provides it to the user's device. The user can then make investment decisions based on this summary. For example, when a user accesses the "Latest News" section of Yahoo! Finance, the server provides a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0610] Furthermore, the server is equipped with an emotion engine that recognizes and evaluates the user's emotional state. The emotion engine analyzes the user's operation history, browsing history, click behavior, browsing time, etc. to evaluate the user's emotional state. For example, the emotion engine analyzes the user's past operation history and determines that the user tends to have positive emotions toward the news that "the Nikkei average rose 2%."

[0611] Based on the user's emotional state analyzed by the emotion engine, the server adjusts the content and display method of the summary information provided to the user's device. It sets the display to prioritize content that the user has a positive feeling about. For example, news about "improving performance of export companies," which the user has a positive feeling about, is displayed with priority along with other related news.

[0612] Prompt Sentence Examples

[0613] "Collect the latest financial news from the Wall Street Journal and extract and summarize the key information."

[0614] "Analyze this news article and identify key keywords."

[0615] "If a user has a positive sentiment towards a particular news item, prioritize showing other related news items."

[0616] In this way, the system of the present invention performs all processes from collecting news data to providing information to the user in one place, and realizes adaptive information provision based on the emotional state of the user.

[0617] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0618] Step 1: Gathering news data

[0619] The server loads the URL or API endpoint of a pre-configured news source, sends a request to the news source's API at the specified time, and retrieves the latest news. Specifically, the server sends a GET request to the API endpoint (e.g., "https: / / example-news-api.com / v2 / latest") at 8:00 every day, and receives the latest news data in JSON format.

[0620] Input: News source URL and API endpoint

[0621] Output: News data in JSON format

[0622] Step 2: Analyzing the news data

[0623] The server analyzes the collected news data using natural language processing technology. An NLP library (e.g., spaCy or NLTK) is used to analyze the content of news articles and identify sentence structure and important keywords. For example, the server analyzes a news article titled "Nikkei Stock Average Rise 2%" and extracts the keywords "Nikkei Stock Average," "Stock Price," and "2% Rise."

[0624] Input: News data in JSON format

[0625] Output: Parsed text data and keywords

[0626] Step 3: Extracting important information

[0627] Based on the analysis results, the server filters and extracts important information necessary for investment decisions. Specifically, the server extracts the specific information that "the main reason for the 2% rise in the Nikkei average is the improved performance of export companies."

[0628] Input: Parsed text data and keywords

[0629] Output: Extracted important information

[0630] Step 4: Generate a summary

[0631] The server uses a summarization algorithm to generate an easy-to-understand summary based on the key information extracted. For example, the server may generate a summary such as, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0632] Input: Extracted sensitive information

[0633] Output: Generated summary

[0634] Step 5: Save the summary

[0635] The server saves the generated summary in a database. Specifically, the database stores the summary content, a link to the original news article, the date and time of generation, etc. For example, the server saves the summary "Nikkei average rises 2%, mainly due to improved performance of export companies" along with the link and the date and time of generation in an SQL database.

[0636] Input: Generated summary

[0637] Output: Summary and detailed information stored in a database

[0638] Step 6: Provide to users

[0639] When a user accesses a specified portal, the server retrieves the latest summary from the database and provides it to the user's device. For example, when a user accesses the "Latest News" section, the server displays a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0640] Input: User request and summary stored in database

[0641] Output: Latest summary displayed on user terminal

[0642] Step 7: Assess your emotional state

[0643] The server uses an emotion engine to analyze the user's operation history, browsing history, click behavior, browsing time, etc., and evaluates the user's emotional state. Specifically, the emotion engine analyzes the user's past operation data and determines that the user tends to have positive emotions toward the news that "the Nikkei average rose 2%."

[0644] Input: User operation history, browsing history, click behavior, browsing time, etc.

[0645] Output: Evaluated user emotional state

[0646] Step 8: Adaptive information provision

[0647] Based on the evaluation results of the emotion engine, the server adjusts the content and display method of the summary information provided to the user's device. For example, it prioritizes the display of news about "improving performance of export companies," which the user has positive feelings about.

[0648] Input: Evaluated emotional state of the user and summaries stored in a database

[0649] Output: Related news displayed in priority order according to the user's emotional state

[0650] (Application example 2)

[0651] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0652] Conventional investment news delivery systems are limited to analyzing collected news information and generating summaries, and do not provide information based on the user's individual emotions or interests. This results in users being exposed to a large amount of unnecessary information, making it difficult to quickly obtain the information necessary for investment decisions. Furthermore, the lack of a mechanism for adaptively providing information based on the user's emotional state makes it difficult to provide personalized investment information.

[0653] The specification processing by the specification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for periodically collecting news information, means for analyzing the collected news information using natural language processing, means for extracting important data from the analyzed information, means for generating a summary based on the extracted information, means for saving the generated summary in a database, means for providing the saved summary to a user terminal, means for recognizing the user's emotional state, and means for adaptively adjusting the information to be provided based on the user's emotional state. This enables users to quickly and efficiently obtain personalized investment information based on their own emotions and interests.

[0654] "Means for regularly collecting news information" refers to a system for obtaining news data from specific news sources at regular intervals.

[0655] "Means for analyzing collected news information using natural language processing" refers to a system that analyzes collected news data using natural language processing technology to understand the content and extract important keywords.

[0656] "Means for extracting important data from analyzed information" refers to a system that filters and extracts information and keywords that are important for investment decisions from news data analyzed using natural language processing.

[0657] A "means for generating a summary based on extracted information" is an algorithm or program that uses key data to summarize information in a way that is easy for users to understand.

[0658] The "means for storing the generated summary in a database" refers to a mechanism for storing the generated summary information in a database so that it can be accessed and searched later.

[0659] The "means for providing the stored summary to the user terminal" is a mechanism for presenting the summary information stored in the database to the user terminal.

[0660] The "means for recognizing the user's emotional state" is a mechanism that analyzes the user's operation history, browsing behavior, etc. to evaluate their current emotions and interests.

[0661] The "means for adaptively adjusting the information provided based on the emotional state" is a mechanism that dynamically changes the information provided and the way it is displayed based on the user's emotional evaluation results.

[0662] The system of the present invention quickly and efficiently collects, analyzes, and summarizes investment-related news, and adaptively provides information based on the user's sentiment. Specific means and processes for implementing this system are described below.

[0663] Gathering news information

[0664] News gathering methods

[0665] The server periodically collects news data from specific news sources, such as major financial news sites and data providers, using HTTP client software such as the requests library in Python.

[0666] News information analysis

[0667] Natural language processing tools

[0668] The server analyzes the collected news data using natural language processing (NLP) technology. NLP uses libraries such as spaCy and NLTK to analyze the content of news articles and identify important keywords and sentence structures. This process converts the content of the news data into an understandable format.

[0669] Extraction of important information

[0670] Important information extraction means

[0671] From the analyzed news data, particularly important information is extracted. This can be done using machine learning algorithms or rule-based analysis methods. For example, the causes of stock price fluctuations and corporate earnings announcements that are important for investment decisions are targeted.

[0672] Generate a summary

[0673] Summary generation means

[0674] The server generates a summary based on the extracted important information using a summary generation algorithm. The algorithm can combine abstractive and extractive summarization techniques. Generative AI models such as GPT-3 are particularly effective.

[0675] Save Summary

[0676] Database storage means

[0677] The generated summaries are then stored in a database, which also contains metadata such as the summary content, a link to the original news article, and the date and time the summary was generated. This process is often performed using a cloud database service such as Amazon RDS.

[0678] Provision to users

[0679] User terminal provision means

[0680] When a user accesses the system through a device (such as a smartphone or PC), the server retrieves the latest summary from the database and provides it to the user, allowing the user to obtain important investment information in a timely manner.

[0681] Introducing the Emotion Engine

[0682] Emotion Engine Means

[0683] An emotion engine is installed on the server to evaluate how users feel about the summary information provided. The emotion engine collects and analyzes data such as user operation history, browsing history, click behavior, and browsing time. Machine learning libraries such as TensorFlow and Keras are used for emotion analysis.

[0684] Adaptive Delivery Methods

[0685] Based on the user's emotional state analyzed by the emotion engine, the server adjusts the content and presentation of the summary information it provides. For example, it may prioritize news stories to which the user has responded positively. This dynamic adjustment is an important feature for providing information optimized for each individual user.

[0686] Specific examples

[0687] 1. News gathering: Access the API endpoint "https: / / example-news-api.com / v2 / latest" to get the latest news data.

[0688] 2. News analysis: Analyze the collected news item "Nikkei average stock price rises 2%" and extract keywords such as "Nikkei average," "stock price," and "2% rise."

[0689] 3. Extraction of important information: Extract important information such as "The main reason for the 2% rise in the Nikkei average stock price is the improved performance of export companies."

[0690] 4. Summary generation: Using GPT-3, generate a summary such as "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0691] 5. Emotion engine: Analyzes the user's past operation history and determines that they tend to have positive emotions toward the news that "Nikkei average rises 2%," and prioritizes displaying content that evokes positive emotions.

[0692] Prompt Sentence Examples

[0693] Prompt for news summary generation:

[0694] News article: "The Nikkei Stock Average rose 2%. This was mainly due to improved performance of export companies."

[0695] Generated summary: "Nikkei average rises 2%, primarily due to improved performance of export companies"

[0696] In this way, the invention can be specifically implemented.

[0697] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0698] Step 1: Get news information

[0699] The server periodically accesses the news API endpoint to retrieve the latest news data. The input is the API endpoint URL, and the output is the retrieved news data (in JSON format). Specifically, it sends an HTTP request using the Python requests library and receives the news data as a response.

[0700] Step 2: Analyzing the news information

[0701] The server analyzes the news data acquired in step 1 using natural language processing (NLP) technology. The input is the acquired news data, and the output is the analyzed important keywords and phrases. Specifically, it uses NLP libraries such as spaCy and NLTK to perform text analysis.

[0702] Step 3: Extracting important information

[0703] The server extracts important information necessary for investment decisions from the data analyzed in step 2. The input is the analyzed keywords and phrases, and the output is the extracted important information. Specifically, it uses machine learning algorithms and rule-based filtering methods to select important information.

[0704] Step 4: Generate a summary

[0705] The server generates a summary using a summary generation algorithm based on the information extracted in step 3. The input is the extracted key information, and the output is the generated summary. Specifically, a generative AI model such as GPT-3 is used to create the summary. An example of a prompt sentence for the generative AI model is, "News article: The Nikkei average stock price rose 2%. This is mainly due to the improved performance of export companies. Generated summary: The Nikkei average rose 2%, mainly due to the improved performance of export companies."

[0706] Step 5: Save the summary

[0707] The server stores the summaries generated in step 4 in a database. The input is the generated summary information, and the output is the summary data stored in the database. Specifically, the server uses an SQL query to insert the summary information into a cloud database such as Amazon RDS.

[0708] Step 6: Deliver to users

[0709] When a user accesses the system from a terminal, the server retrieves the latest summary from the database and provides it to the user's terminal. The input is the user's access request, and the output is the displayed summary information. Specifically, it receives the user's request, retrieves the summary information from the database using an SQL query, and returns it as an HTTP response.

[0710] Step 7: Recognizing your emotional state

[0711] The server analyzes the user's operation history and browsing behavior to evaluate their emotional state. The input is the user's past operation data, and the output is the evaluation result of the user's emotional state. Specifically, emotion analysis is performed using machine learning libraries such as TensorFlow and Keras.

[0712] Step 8: Adaptively adjusting information provision

[0713] The server adjusts the content and presentation of the information it provides based on the emotion evaluation results obtained in step 7. The input is the evaluation result of the emotional state, and the output is the adjusted summary information. Specifically, it reorganizes the list of high-priority information and applies an algorithm to display it in a form optimized for the user.

[0714] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0715] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0716] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0717] [Third embodiment]

[0718] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0719] 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.

[0720] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0721] 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.

[0722] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0723] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0724] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0725] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0726] 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.

[0727] 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.

[0728] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0729] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0730] The system of the present invention collects, analyzes, summarizes, and provides investment-related news quickly and efficiently. To implement this system, the following specific forms are considered and implemented.

[0731] Gathering news information

[0732] News gathering methods

[0733] The server pre-configures the URLs and API endpoints of specific news sources (e.g., major financial news sites or data providers) and periodically collects news data based on these. For example, the server retrieves the latest news from the Wall Street Journal or Reuters APIs at 8:00 every day.

[0734] Example: The server accesses the API endpoint "https: / / example-news-api.com / v2 / latest" to retrieve the latest news data in JSON format.

[0735] News information analysis

[0736] Natural language processing tools

[0737] The server analyzes the collected news data using natural language processing (NLP) techniques. NLP libraries (e.g., spaCy and NLTK) are used to analyze the content of news articles and identify important keywords and sentence structures.

[0738] Example: The server analyzes the news article "Nikkei average stock price rises 2%" and extracts keywords such as "Nikkei average," "stock price," and "2% rise."

[0739] Extraction of important information

[0740] Important information extraction means

[0741] Based on the analysis results, the server filters and extracts important information necessary for investment decisions (for example, causes of stock price fluctuations, corporate earnings announcements, etc.).

[0742] Example: From the news data analyzed by the server, specific information such as "The main reason for the 2% rise in the Nikkei average stock price is the improved performance of export companies" is extracted.

[0743] Generate a summary

[0744] Summary generation means

[0745] Based on the key information extracted by the server, a summarization algorithm is used to generate a summary that is easy for investors to understand. Abstractive summarization and extractive summarization techniques can be combined.

[0746] Example: The server generates a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0747] Save Summary

[0748] Database storage means

[0749] The server stores the generated summaries in a database, including the summary content, a link to the original news article, and the date and time the summary was generated.

[0750] Example: The server stores in an SQL database the summary "The Nikkei average rose 2%, primarily due to improved performance of export companies" along with further information.

[0751] Provision to users

[0752] User terminal provision means

[0753] When a user accesses the Yahoo Finance portal, the server retrieves the latest summary from the database and provides it to the user's device, allowing the user to make investment decisions based on it.

[0754] Example: When a user visits the "Latest News" section of Yahoo Finance, they are presented with a server-provided summary that reads, "Nikkei average rises 2%, primarily due to improved performance of export companies."

[0755] The system of the present invention includes these means and is capable of collecting, analyzing, summarizing and providing investment-related news quickly and efficiently, thereby greatly assisting investors in obtaining information.

[0756] The processing flow will be explained below.

[0757] Specific processing steps of the news gathering and analysis system

[0758] Step 1:

[0759] The server configures the URL and API endpoint of the news source, for example, registering the endpoint of a particular financial news site or data provider.

[0760] Step 2:

[0761] The server retrieves news data from multiple news sources according to a pre-defined schedule, for example, calling the API every day at 8:00 to retrieve the latest news.

[0762] Step 3:

[0763] The news data retrieved by the server is preprocessed, specifically by extracting text from HTML content and converting it into a parseable format.

[0764] Step 4:

[0765] The server uses natural language processing (NLP) techniques to analyze the preprocessed news data, using NLP libraries to analyze sentence structure and extract keywords.

[0766] Step 5:

[0767] The server then uses the analysis results to extract key information from news articles, filtering the information based on specific keywords such as "stock prices," "earnings announcements," and "market trends."

[0768] Step 6:

[0769] The server organizes the extracted key information and applies a summary generation algorithm to generate summaries that are easy for investors to understand. Abstractive summaries and extractive summaries are used.

[0770] Step 7:

[0771] The server stores the generated summary in a database, including the summary content, the URL of the original article, and the date and time the summary was generated.

[0772] Step 8:

[0773] A user visits the Yahoo! Finance portal and requests to view breaking news.

[0774] Step 9:

[0775] The server retrieves the latest summary data from the database and sends it to the user terminal.

[0776] Step 10:

[0777] The user's device displays the received summary data on the Yahoo Finance interface, allowing the user to view the information.

[0778] Example 1

[0779] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0780] Because news information is vast and scattered, investors do not have time to research and analyze it individually, making it difficult to quickly and efficiently obtain important news information. Furthermore, analyzing news information and generating summaries requires advanced technology and knowledge, and doing this manually is labor-intensive. Therefore, there is a need for a system that can quickly and accurately provide the important news information investors need.

[0781] 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.

[0782] In this invention, the server includes means for periodically collecting news information, means for preprocessing the collected news information, means for analyzing the preprocessed news information using natural language processing, means for extracting important data from the analyzed information, means for generating summaries based on the extracted important data, means for storing the generated summaries in a database, and means for providing the stored summaries to a user terminal, thereby enabling investors to quickly and efficiently obtain the latest important news information and make investment decisions.

[0783] "Means for regularly collecting news information" refers to a function that allows you to set the URL or API endpoint of a specific news source and obtain news data daily or at a fixed time.

[0784] "Means for preprocessing collected news information" refers to a function for removing unnecessary information from acquired news data and converting it into a format that can be analyzed.

[0785] "Means for analyzing preprocessed news information using natural language processing" refers to a function that uses natural language processing techniques and libraries to analyze preprocessed news text and identify important keywords and sentence structures.

[0786] "Means of extracting important data from analyzed information" refers to a function that filters and selects important information necessary for investment decisions from the data obtained as a result of natural language processing.

[0787] The "means for generating a summary based on the extracted important data" is a function that uses an algorithm to create a summary based on the extracted important information in a form that is easy for investors to understand.

[0788] "Means for saving the generated summary in a database" refers to a function that saves the generated summary and its related information in a database so that it can be referenced or searched later.

[0789] The "means for providing the stored summary to the user terminal" is a function for retrieving the latest summary from the database and displaying it on the user terminal.

[0790] The system of the present invention collects, analyzes, summarizes, and provides investment-related news quickly and efficiently. When this system is implemented, the server, terminals, and users work together to perform specific processes. A specific embodiment of this system is described below.

[0791] Gathering news information

[0792] The server configures the URL and API endpoint of a specific news source. News sources include major financial news sites and data providers. For example, the server accesses the API endpoint of a financial news site every day at 8:00 AM to retrieve the latest news data. Specifically, consider the server sending a GET request to the API endpoint "https: / / example-news-api.com / v2 / latest" to retrieve the news data in JSON format.

[0793] Preprocessing of news information

[0794] The news data acquired by the server is preprocessed before analysis. This preprocessing involves extracting only the news text and removing HTML tags and certain unnecessary characters. This converts the data into a format suitable for natural language processing.

[0795] News information analysis

[0796] The server analyzes the preprocessed news data using natural language processing (NLP) technology. SpaCy or NLTK can be used as an NLP library. NLP technology analyzes the content of news articles and identifies important keywords and sentence structures. For example, the server analyzes the news article "Nikkei Stock Average Rise 2%" and extracts keywords such as "Nikkei Stock Average," "Stock Price," and "2% Rise."

[0797] Extraction of important information

[0798] Based on the analysis results, the server filters and extracts important information necessary for investment decisions. Important information includes the causes of stock price fluctuations and corporate earnings announcements. For example, the server extracts specific information such as "The main reason for the 2% rise in the Nikkei average is the improved performance of export companies."

[0799] Generate a summary

[0800] Based on the key information extracted by the server, a summarization algorithm is used to generate summaries that are easy for investors to understand. Abstractive and extractive summarization techniques are used to generate summaries. For example, the server might generate a summary such as, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0801] Save Summary

[0802] The server saves the generated summary in a database. The data saved in the database includes the summary content, a link to the original news article, the date and time the summary was generated, etc. Specifically, the server saves the summary "The Nikkei average rose 2%, mainly due to improved performance of export companies" in an SQL database.

[0803] Provision to users

[0804] When a user accesses a financial portal site, the server retrieves the latest summary from the database and provides it to the user's device. The user can make investment decisions based on this summary. For example, when a user accesses the "Latest News" section of a financial portal site, a summary provided by the server is displayed: "The Nikkei average rose 2%, mainly due to improved performance of export companies."

[0805] Example prompts for generative AI models

[0806] By inputting the following prompt sentences into the generative AI model, the program's processing can be made easier to understand.

[0807] Example: "Describe a program that retrieves the latest investment-related news from financial news sites, uses natural language processing techniques to extract key points, and generates summaries that are easy for investors to understand."

[0808] This will enable investors to obtain the necessary information quickly and efficiently and make investment decisions.

[0809] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0810] Step 1:

[0811] The server loads the news source configuration. The server reads a previously saved configuration file and obtains the URLs and API endpoints of multiple news sources. The input is the contents of the configuration file, and the output is a list of news sources. Specifically, the server parses the "config / news_sources.json" file and stores the endpoints of each news source in a list.

[0812] Step 2:

[0813] The server periodically collects news data. Using the endpoints of the loaded news sources, the server sends API requests daily or at regular intervals to obtain the latest news data. The input is a list of news sources, and the output is news data in JSON format. Specifically, the server sends a GET request to each API endpoint at 8:00 every day and saves the received data as JSON.

[0814] Step 3:

[0815] The server preprocesses the acquired news data. The server removes unnecessary information from the acquired news data and cleans up the text data. The input is news data in JSON format, and the output is text data suitable for analysis. Specifically, the server parses the JSON data to extract the news text and removes HTML tags and unnecessary characters.

[0816] Step 4:

[0817] The server analyzes the preprocessed news data using natural language processing technology. The server uses an NLP library (for example, spaCy or NLTK) to analyze the content of the news article and identify important keywords and sentence structure. The input is the preprocessed text data, and the output is the analysis results (keywords and sentence structure). Specifically, the server analyzes the news article "Nikkei Stock Average Rise 2%" and extracts keywords such as "Nikkei Stock Average," "Stock Price," and "2% Rise."

[0818] Step 5:

[0819] The server extracts important information from the analysis results. Based on the results of natural language processing, the server filters and extracts important information necessary for investment decisions. The input is the analysis results, and the output is important information. Specifically, the server extracts information such as "The main reason for the 2% rise in the Nikkei average is the improved performance of export companies."

[0820] Step 6:

[0821] The server generates a summary based on the important information. Using a summarization algorithm, the server extracts the important information and creates a summary in a format that is easy for investors to understand. The input is the important information, and the output is the generated summary. Specifically, the server generates the summary, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0822] Step 7:

[0823] The server saves the generated summary in a database. The server stores the generated summary and its related information in a database. The input is the generated summary and its related information, and the output is the result saved in the database. Specifically, the server saves the summary "Nikkei average rises 2%, mainly due to improved performance of export companies" in an SQL database.

[0824] Step 8:

[0825] When a user accesses the portal site, the server provides the latest summary to the user's device. The server retrieves the latest summary from the database and sends it to the user's device. The input is the user's request, and the output is the summary data. In concrete terms, when a user accesses the "Latest News" section of the portal site, the server displays the summary "Nikkei average rises 2%, mainly due to improved performance of export companies."

[0826] (Application example 1)

[0827] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0828] There is a need to quickly and efficiently collect, analyze, summarize, and provide investors with a vast amount of investment-related news information. However, performing these tasks manually takes time and effort, and there is a risk of missing important timing for investment decisions. In addition, technology is needed to unify information management from different news sources and appropriately extract and summarize important information.

[0829] 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.

[0830] In this invention, the server includes means for periodically collecting news information, means for analyzing the collected news information using natural language processing, and means for extracting important data from the analyzed information, thereby enabling investment-related news to be collected, analyzed, and summarized quickly and efficiently, and provided to investors in a timely manner.

[0831] "News information" refers to current events and information collected from specific news sources.

[0832] "Periodic collection means" refers to a function that automatically obtains news information at specific time intervals.

[0833] "Means of analysis using natural language processing" refers to a technology that converts text data into a form that a computer can understand and analyzes the content.

[0834] "Means for extracting important data" refers to the function of identifying and extracting important information necessary for investment decisions from analyzed news information.

[0835] "Means for generating a summary" refers to a function that concisely summarizes the content based on the extracted important information.

[0836] "Means for storing in a database" refers to the function of storing the generated summaries in an organized manner so that they can be searched and referenced later.

[0837] "Means for providing to user terminal" refers to the function of displaying and transmitting the stored summary information to the user's device.

[0838] "Means for users to access summaries via a web application" means the ability for users to view news summaries on a web platform via the Internet.

[0839] A system for implementing the present invention includes multiple means for quickly and efficiently collecting, analyzing, summarizing, storing, and presenting investment-related news.

[0840] Gathering news information

[0841] The server pre-configures the API endpoint of a specific news source and periodically collects news data. For example, this process can be performed using the Python Requests library. For example, the server periodically accesses the news API to retrieve the latest news data. The collected data is stored in JSON format.

[0842] News information analysis

[0843] The server analyzes the collected news data using natural language processing (NLP) techniques, such as the spaCy library in Python, to identify important keywords and sentence structures from the content of news articles.

[0844] Extraction of important information

[0845] The server then uses the analysis results to extract key data needed to make investment decisions. This is done through keyword matching and rule-based filtering. For example, keywords such as "stock price" and "rise" are extracted from news articles.

[0846] Generate a summary

[0847] The server uses a summarization algorithm to generate a summary based on the extracted important information. This process uses the Python transformers library (e.g., the BERT model). For example, a summary such as "Stock prices rose 2% due to improved performance of export companies" is generated.

[0848] Save Summary

[0849] The server stores the generated summary in an SQL database. The stored data includes the summary content, a link to the original news article, the date and time the summary was generated, etc. For example, a summary such as "Stock prices rose 2% due to improved performance of export companies" is stored in the database.

[0850] Providing a summary

[0851] Users can access news summaries through a web application, which uses the Python Flask framework. For example, when a user accesses the web application, the server retrieves the latest summaries from the database and provides them to the user.

[0852] Hardware and software used

[0853] Hardware: Servers, database servers, web servers

[0854] Software: Python, Flask, spaCy, transformers, sqlite3

[0855] Prompt Sentence Examples

[0856] "Implement a system to collect the latest investment news, analyze it, extract key information, and generate summaries to provide to users."

[0857] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0858] Step 1:

[0859] The server accesses the news source's API endpoint and periodically retrieves news information. The input is the news source's API endpoint, and the output is news data in JSON format. The server uses Python's requests library to send requests to the news API and retrieves the news data received as a response.

[0860] Step 2:

[0861] The server analyzes the acquired news data using natural language processing (NLP) techniques. The input is news data in JSON format, and the output is a list of important keywords contained in the news articles. The server uses the Python spaCy library to extract important keywords and sentence structure from the content of the news articles. Specifically, it analyzes the text of the news articles and performs tokenization and part-of-speech tagging.

[0862] Step 3:

[0863] The server extracts key data based on the analysis results. The input is the analysis results of the news article, and the output is a list of key information needed to make investment decisions. Through keyword matching and rule-based filtering, the server identifies and extracts information containing key keywords such as "stock price," "rise," and "performance."

[0864] Step 4:

[0865] The server uses a summarization algorithm to generate a summary based on the extracted key information. The input is a list of key information, and the output is the summarized text. The server uses the Python transformers library and a generative AI model (e.g., the BERT model) to generate a concise summary of the key information.

[0866] Step 5:

[0867] The server stores the generated summaries in an SQL database. The input is the summarized text, and the output is the records stored in the database. The server uses Python's sqlite3 library to store the summary content, a link to the original news article, and the date and time the summary was generated in the database.

[0868] Step 6:

[0869] A user accesses the latest news summaries through a web application. The input is an access request from the user, and the output is a user interface that displays the latest news summaries. The server uses the Python Flask framework to retrieve the latest summaries from the database and provide them to the user's device. Specifically, when a user sends a request through the web application, the server queries the database, retrieves the latest summaries, and displays them to the user.

[0870] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0871] The system of the present invention not only quickly and efficiently collects, analyzes, summarizes, and provides investment-related news, but also recognizes user emotions and adaptively provides information based on those emotions. To implement this system, the following specific forms are considered and implemented.

[0872] Gathering news information

[0873] News gathering methods

[0874] The server pre-configures the URLs and API endpoints of specific news sources (e.g., major financial news sites or data providers) and periodically collects news data based on these. For example, the server retrieves the latest news from the Wall Street Journal or Reuters APIs at 8:00 every day.

[0875] Example: The server accesses the API endpoint "https: / / example-news-api.com / v2 / latest" to retrieve the latest news data in JSON format.

[0876] News information analysis

[0877] Natural language processing tools

[0878] The server analyzes the collected news data using natural language processing (NLP) techniques. NLP libraries (e.g., spaCy and NLTK) are used to analyze the content of news articles and identify important keywords and sentence structures.

[0879] Example: The server analyzes the news article "Nikkei average stock price rises 2%" and extracts keywords such as "Nikkei average," "stock price," and "2% rise."

[0880] Extraction of important information

[0881] Important information extraction means

[0882] Based on the analysis results, the server filters and extracts important information necessary for investment decisions (for example, causes of stock price fluctuations, corporate earnings announcements, etc.).

[0883] Example: From the news data analyzed by the server, specific information such as "The main reason for the 2% rise in the Nikkei average stock price is the improved performance of export companies" is extracted.

[0884] Generate a summary

[0885] Summary generation means

[0886] Based on the key information extracted by the server, a summarization algorithm is used to generate a summary that is easy for investors to understand. Abstractive summarization and extractive summarization techniques can be combined.

[0887] Example: The server generates a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0888] Save Summary

[0889] Database storage means

[0890] The server stores the generated summaries in a database, including the summary content, a link to the original news article, and the date and time the summary was generated.

[0891] Example: The server stores in an SQL database the summary "The Nikkei average rose 2%, primarily due to improved performance of export companies" along with further information.

[0892] Provision to users

[0893] User terminal provision means

[0894] When a user accesses the Yahoo Finance portal, the server retrieves the latest summary from the database and provides it to the user's device, allowing the user to make investment decisions based on it.

[0895] Example: When a user visits the "Latest News" section of Yahoo Finance, they are presented with a server-provided summary that reads, "Nikkei average rises 2%, primarily due to improved performance of export companies."

[0896] Introducing the Emotion Engine

[0897] Emotion Engine Means

[0898] An emotion engine is installed on the server to recognize how the user feels about the provided summary information. The emotion engine analyzes the user's operation history, browsing history, click behavior, browsing time, etc. to evaluate the user's emotional state.

[0899] Example: The emotion engine analyzes the user's past behavior history and determines that they tend to have positive emotions toward the news that "the Nikkei average has risen 2%."

[0900] Adaptive Delivery Methods

[0901] Based on the user's emotional state analyzed by the emotion engine, the server adjusts the content and presentation of the summary information provided to the user, for example, by prioritizing the display of content that the user feels positive about.

[0902] Example: Prioritize displaying news about "improved performance of export companies," which users have positive feelings about, along with other related news.

[0903] The system of the present invention includes these means and quickly and efficiently collects, analyzes, summarizes, and provides investment-related news, and also adaptively provides information taking into account the user's emotional state, thereby realizing more personalized investment information provision.

[0904] The processing flow will be explained below.

[0905] Specific processing steps of the news gathering and analysis system (including the sentiment engine)

[0906] Step 1:

[0907] The server configures the URL and API endpoint of the news source, for example, registering the endpoint of a particular financial news site or data provider.

[0908] Step 2:

[0909] The server retrieves news data from multiple news sources according to a pre-defined schedule, for example, calling the API every day at 8:00 to retrieve the latest news.

[0910] Step 3:

[0911] The news data retrieved by the server is preprocessed, specifically by extracting text from HTML content and converting it into a parseable format.

[0912] Step 4:

[0913] The server uses natural language processing (NLP) techniques to analyze the preprocessed news data, using NLP libraries to analyze sentence structure and extract keywords.

[0914] Step 5:

[0915] The server then uses the analysis results to extract key information from news articles, filtering the information based on specific keywords such as "stock prices," "earnings announcements," and "market trends."

[0916] Step 6:

[0917] The server organizes the extracted key information and applies a summary generation algorithm to generate summaries that are easy for investors to understand. Abstractive summaries and extractive summaries are used.

[0918] Step 7:

[0919] The server stores the generated summary in a database, including the summary content, the URL of the original article, and the date and time the summary was generated.

[0920] Step 8:

[0921] A user visits the Yahoo! Finance portal and requests to view breaking news.

[0922] Step 9:

[0923] The server retrieves the latest summary data from the database and sends it to the user terminal.

[0924] Step 10:

[0925] The user's device displays the received summary data on the Yahoo Finance interface, allowing the user to view the information.

[0926] Specific steps for implementing the Emotion Engine

[0927] Step 11:

[0928] The server collects the user's operation history, browsing history, click behavior, browsing time, etc. using an emotion engine.

[0929] Step 12:

[0930] The server analyzes the collected user data and evaluates their emotional state, for example classifying them as positive, negative, or neutral.

[0931] Step 13:

[0932] The server adjusts the content and presentation of the summary information provided based on the user's emotional state, giving priority to content that the user has a positive feeling about.

[0933] Step 14:

[0934] The user views the summary information provided preferentially and accumulates their operation history. The server then analyzes this data using the emotion engine and reflects it in the next information provided.

[0935] Specific examples

[0936] Examples:

[0937] The server retrieves news data such as "Nikkei Stock Average rises 2%" from the financial news API. The server performs NLP analysis to extract keywords such as "Nikkei Stock Average," "stock price," and "2% rise," and applies a summary generation algorithm to generate a summary such as "Nikkei Stock Average rises 2%, primarily due to improved performance of export companies." The server stores this summary in a database, and the user accesses Yahoo! Finance. The server then analyzes the user's past operation history using an emotion engine, and prioritizes displaying news with positive sentiment. This provides users with information they need to make immediate investment decisions.

[0938] Example 2

[0939] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0940] Conventional news information gathering and analysis systems collect, analyze, and summarize news, but they have the problem of being unable to provide information that takes into account the user's emotional state. As a result, it is difficult to provide the information that the user actually needs at the appropriate time, and this can result in excessive or insufficient provision of information.

[0941] 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 means for periodically collecting news data, means for analyzing the collected news data using natural language processing technology, means for extracting important information from the analyzed data, means for generating a summary based on the extracted information, means for saving the generated summary in a storage device, means for providing the saved summary to a user terminal, means for recognizing and evaluating the emotional state of a user, and means for adaptively providing information based on the evaluation result. This makes it possible to provide information adaptively taking the emotional state of a user into consideration.

[0942] "News data" is text data about current events and information, obtained from multiple news sources.

[0943] "Natural language processing technology" is a technology for processing human language using a computer, and involves analyzing text and extracting keywords.

[0944] "Storage" refers to a storage device or database for storing generated summaries.

[0945] A "usage terminal" is a device such as a computer, smartphone, or tablet that a user uses to receive information.

[0946] "User's emotional state" refers to the positive or negative emotional state that a user feels when viewing news information.

[0947] "Adaptive information provision" refers to providing information in a form optimized for a user's emotional state, taking into account that user's emotional state.

[0948] "News source" refers to the source of information that provides news data, including online newspapers, news sites, API services, etc.

[0949] This invention is a system that not only quickly and efficiently collects, analyzes, summarizes, and provides investment-related news, but also recognizes the user's emotional state and adaptively provides information based on that.

[0950] The server periodically collects news data using pre-configured news source URLs or API endpoints. For example, it uses the API endpoints of major financial news sites and data providers. Specifically, the server accesses the Wall Street Journal or Reuters API endpoint "https: / / example-news-api.com / v2 / latest" at 8:00 every day, and retrieves the latest news in JSON format.

[0951] Next, the server analyzes the collected news data using natural language processing (NLP) technology. NLP libraries such as spaCy and NLTK are used to analyze the content of news articles and identify important keywords and sentence structures. For example, the server analyzes the news article "Nikkei Stock Average Rise 2%" and extracts the keywords "Nikkei Stock Average," "Stock Price," and "Rise 2%."

[0952] Next, the server extracts important information based on the analysis results. It filters and extracts information necessary for investment decisions (e.g., causes of stock price fluctuations, corporate earnings announcements, etc.). For example, the server might extract specific information such as, "The main reason the Nikkei average rose 2% is the improved performance of export companies."

[0953] The server then uses a summarization algorithm based on the key information extracted to generate a summary that is easy for investors to understand. Abstractive and extractive summarization techniques can be combined. For example, the server generates a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0954] The server stores the generated summary in a storage device, such as an SQL database. The stored data includes the summary content, a link to the original news article, the date and time the summary was generated, etc. For example, the server stores the summary "The Nikkei average rose 2%, mainly due to improved performance of export companies" and its detailed information in the SQL database.

[0955] When a user accesses the Yahoo! Finance portal, the server retrieves the latest summary from the database and provides it to the user's device. The user can then make investment decisions based on this summary. For example, when a user accesses the "Latest News" section of Yahoo! Finance, the server provides a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0956] Furthermore, the server is equipped with an emotion engine that recognizes and evaluates the user's emotional state. The emotion engine analyzes the user's operation history, browsing history, click behavior, browsing time, etc. to evaluate the user's emotional state. For example, the emotion engine analyzes the user's past operation history and determines that the user tends to have positive emotions toward the news that "the Nikkei average rose 2%."

[0957] Based on the user's emotional state analyzed by the emotion engine, the server adjusts the content and display method of the summary information provided to the user's device. It sets the display to prioritize content that the user has a positive feeling about. For example, news about "improving performance of export companies," which the user has a positive feeling about, is displayed with priority along with other related news.

[0958] Prompt Sentence Examples

[0959] "Collect the latest financial news from the Wall Street Journal and extract and summarize the key information."

[0960] "Analyze this news article and identify key keywords."

[0961] "If a user has a positive sentiment towards a particular news item, prioritize showing other related news items."

[0962] In this way, the system of the present invention performs all processes from collecting news data to providing information to the user in one place, and realizes adaptive information provision based on the emotional state of the user.

[0963] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0964] Step 1: Gathering news data

[0965] The server loads the URL or API endpoint of a pre-configured news source, sends a request to the news source's API at the specified time, and retrieves the latest news. Specifically, the server sends a GET request to the API endpoint (e.g., "https: / / example-news-api.com / v2 / latest") at 8:00 every day, and receives the latest news data in JSON format.

[0966] Input: News source URL and API endpoint

[0967] Output: News data in JSON format

[0968] Step 2: Analyzing the news data

[0969] The server analyzes the collected news data using natural language processing technology. An NLP library (e.g., spaCy or NLTK) is used to analyze the content of news articles and identify sentence structure and important keywords. For example, the server analyzes a news article titled "Nikkei Stock Average Rise 2%" and extracts the keywords "Nikkei Stock Average," "Stock Price," and "2% Rise."

[0970] Input: News data in JSON format

[0971] Output: Parsed text data and keywords

[0972] Step 3: Extracting important information

[0973] Based on the analysis results, the server filters and extracts important information necessary for investment decisions. Specifically, the server extracts the specific information that "the main reason for the 2% rise in the Nikkei average is the improved performance of export companies."

[0974] Input: Parsed text data and keywords

[0975] Output: Extracted important information

[0976] Step 4: Generate a summary

[0977] The server uses a summarization algorithm to generate an easy-to-understand summary based on the key information extracted. For example, the server may generate a summary such as, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0978] Input: Extracted sensitive information

[0979] Output: Generated summary

[0980] Step 5: Save the summary

[0981] The server saves the generated summary in a database. Specifically, the database stores the summary content, a link to the original news article, the date and time of generation, etc. For example, the server saves the summary "Nikkei average rises 2%, mainly due to improved performance of export companies" along with the link and the date and time of generation in an SQL database.

[0982] Input: Generated summary

[0983] Output: Summary and detailed information stored in a database

[0984] Step 6: Provide to users

[0985] When a user accesses a specified portal, the server retrieves the latest summary from the database and provides it to the user's device. For example, when a user accesses the "Latest News" section, the server displays a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[0986] Input: User request and summary stored in database

[0987] Output: Latest summary displayed on user terminal

[0988] Step 7: Assess your emotional state

[0989] The server uses an emotion engine to analyze the user's operation history, browsing history, click behavior, browsing time, etc., and evaluates the user's emotional state. Specifically, the emotion engine analyzes the user's past operation data and determines that the user tends to have positive emotions toward the news that "the Nikkei average rose 2%."

[0990] Input: User operation history, browsing history, click behavior, browsing time, etc.

[0991] Output: Evaluated user emotional state

[0992] Step 8: Adaptive information provision

[0993] Based on the evaluation results of the emotion engine, the server adjusts the content and display method of the summary information provided to the user's device. For example, it prioritizes the display of news about "improving performance of export companies," which the user has positive feelings about.

[0994] Input: Evaluated emotional state of the user and summaries stored in a database

[0995] Output: Related news displayed in priority order according to the user's emotional state

[0996] (Application example 2)

[0997] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0998] Conventional investment news delivery systems are limited to analyzing collected news information and generating summaries, and do not provide information based on the user's individual emotions or interests. This results in users being exposed to a large amount of unnecessary information, making it difficult to quickly obtain the information necessary for investment decisions. Furthermore, the lack of a mechanism for adaptively providing information based on the user's emotional state makes it difficult to provide personalized investment information.

[0999] The specification processing by the specification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for periodically collecting news information, means for analyzing the collected news information using natural language processing, means for extracting important data from the analyzed information, means for generating a summary based on the extracted information, means for saving the generated summary in a database, means for providing the saved summary to a user terminal, means for recognizing the user's emotional state, and means for adaptively adjusting the information to be provided based on the user's emotional state. This enables users to quickly and efficiently obtain personalized investment information based on their own emotions and interests.

[1000] "Means for regularly collecting news information" refers to a system for obtaining news data from specific news sources at regular intervals.

[1001] "Means for analyzing collected news information using natural language processing" refers to a system that analyzes collected news data using natural language processing technology to understand the content and extract important keywords.

[1002] "Means for extracting important data from analyzed information" refers to a system that filters and extracts information and keywords that are important for investment decisions from news data analyzed using natural language processing.

[1003] A "means for generating a summary based on extracted information" is an algorithm or program that uses key data to summarize information in a way that is easy for users to understand.

[1004] The "means for storing the generated summary in a database" refers to a mechanism for storing the generated summary information in a database so that it can be accessed and searched later.

[1005] The "means for providing the stored summary to the user terminal" is a mechanism for presenting the summary information stored in the database to the user terminal.

[1006] The "means for recognizing the user's emotional state" is a mechanism that analyzes the user's operation history, browsing behavior, etc. to evaluate their current emotions and interests.

[1007] The "means for adaptively adjusting the information provided based on the emotional state" is a mechanism that dynamically changes the information provided and the way it is displayed based on the user's emotional evaluation results.

[1008] The system of the present invention quickly and efficiently collects, analyzes, and summarizes investment-related news, and adaptively provides information based on the user's sentiment. Specific means and processes for implementing this system are described below.

[1009] Gathering news information

[1010] News gathering methods

[1011] The server periodically collects news data from specific news sources, such as major financial news sites and data providers, using HTTP client software such as the requests library in Python.

[1012] News information analysis

[1013] Natural language processing tools

[1014] The server analyzes the collected news data using natural language processing (NLP) technology. NLP uses libraries such as spaCy and NLTK to analyze the content of news articles and identify important keywords and sentence structures. This process converts the content of the news data into an understandable format.

[1015] Extraction of important information

[1016] Important information extraction means

[1017] From the analyzed news data, particularly important information is extracted. This can be done using machine learning algorithms or rule-based analysis methods. For example, the causes of stock price fluctuations and corporate earnings announcements that are important for investment decisions are targeted.

[1018] Generate a summary

[1019] Summary generation means

[1020] The server generates a summary based on the extracted important information using a summary generation algorithm. The algorithm can combine abstractive and extractive summarization techniques. Generative AI models such as GPT-3 are particularly effective.

[1021] Save Summary

[1022] Database storage means

[1023] The generated summaries are then stored in a database, which also contains metadata such as the summary content, a link to the original news article, and the date and time the summary was generated. This process is often performed using a cloud database service such as Amazon RDS.

[1024] Provision to users

[1025] User terminal provision means

[1026] When a user accesses the system through a device (such as a smartphone or PC), the server retrieves the latest summary from the database and provides it to the user, allowing the user to obtain important investment information in a timely manner.

[1027] Introducing the Emotion Engine

[1028] Emotion Engine Means

[1029] An emotion engine is installed on the server to evaluate how users feel about the summary information provided. The emotion engine collects and analyzes data such as user operation history, browsing history, click behavior, and browsing time. Machine learning libraries such as TensorFlow and Keras are used for emotion analysis.

[1030] Adaptive Delivery Methods

[1031] Based on the user's emotional state analyzed by the emotion engine, the server adjusts the content and presentation of the summary information it provides. For example, it may prioritize news stories to which the user has responded positively. This dynamic adjustment is an important feature for providing information optimized for each individual user.

[1032] Specific examples

[1033] 1. News gathering: Access the API endpoint "https: / / example-news-api.com / v2 / latest" to get the latest news data.

[1034] 2. News analysis: Analyze the collected news item "Nikkei average stock price rises 2%" and extract keywords such as "Nikkei average," "stock price," and "2% rise."

[1035] 3. Extraction of important information: Extract important information such as "The main reason for the 2% rise in the Nikkei average stock price is the improved performance of export companies."

[1036] 4. Summary generation: Using GPT-3, generate a summary such as "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[1037] 5. Emotion engine: Analyzes the user's past operation history and determines that they tend to have positive emotions toward the news that "Nikkei average rises 2%," and prioritizes displaying content that evokes positive emotions.

[1038] Prompt Sentence Examples

[1039] Prompt for news summary generation:

[1040] News article: "The Nikkei Stock Average rose 2%. This was mainly due to improved performance of export companies."

[1041] Generated summary: "Nikkei average rises 2%, primarily due to improved performance of export companies"

[1042] In this way, the invention can be specifically implemented.

[1043] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1044] Step 1: Get news information

[1045] The server periodically accesses the news API endpoint to retrieve the latest news data. The input is the API endpoint URL, and the output is the retrieved news data (in JSON format). Specifically, it sends an HTTP request using the Python requests library and receives the news data as a response.

[1046] Step 2: Analyzing the news information

[1047] The server analyzes the news data acquired in step 1 using natural language processing (NLP) technology. The input is the acquired news data, and the output is the analyzed important keywords and phrases. Specifically, it uses NLP libraries such as spaCy and NLTK to perform text analysis.

[1048] Step 3: Extracting important information

[1049] The server extracts important information necessary for investment decisions from the data analyzed in step 2. The input is the analyzed keywords and phrases, and the output is the extracted important information. Specifically, it uses machine learning algorithms and rule-based filtering methods to select important information.

[1050] Step 4: Generate a summary

[1051] The server generates a summary using a summary generation algorithm based on the information extracted in step 3. The input is the extracted key information, and the output is the generated summary. Specifically, a generative AI model such as GPT-3 is used to create the summary. An example of a prompt sentence for the generative AI model is, "News article: The Nikkei average stock price rose 2%. This is mainly due to the improved performance of export companies. Generated summary: The Nikkei average rose 2%, mainly due to the improved performance of export companies."

[1052] Step 5: Save the summary

[1053] The server stores the summaries generated in step 4 in a database. The input is the generated summary information, and the output is the summary data stored in the database. Specifically, the server uses an SQL query to insert the summary information into a cloud database such as Amazon RDS.

[1054] Step 6: Deliver to users

[1055] When a user accesses the system from a terminal, the server retrieves the latest summary from the database and provides it to the user's terminal. The input is the user's access request, and the output is the displayed summary information. Specifically, it receives the user's request, retrieves the summary information from the database using an SQL query, and returns it as an HTTP response.

[1056] Step 7: Recognizing your emotional state

[1057] The server analyzes the user's operation history and browsing behavior to evaluate their emotional state. The input is the user's past operation data, and the output is the evaluation result of the user's emotional state. Specifically, emotion analysis is performed using machine learning libraries such as TensorFlow and Keras.

[1058] Step 8: Adaptively adjusting information provision

[1059] The server adjusts the content and presentation of the information it provides based on the emotion evaluation results obtained in step 7. The input is the evaluation result of the emotional state, and the output is the adjusted summary information. Specifically, it reorganizes the list of high-priority information and applies an algorithm to display it in a form optimized for the user.

[1060] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1061] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1062] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1063] [Fourth embodiment]

[1064] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1065] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1066] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1067] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1068] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1069] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1070] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1071] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1072] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1073] 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.

[1074] 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.

[1075] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1076] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1077] The system of the present invention collects, analyzes, summarizes, and provides investment-related news quickly and efficiently. To implement this system, the following specific forms are considered and implemented.

[1078] Gathering news information

[1079] News gathering methods

[1080] The server pre-configures the URLs and API endpoints of specific news sources (e.g., major financial news sites or data providers) and periodically collects news data based on these. For example, the server retrieves the latest news from the Wall Street Journal or Reuters APIs at 8:00 every day.

[1081] Example: The server accesses the API endpoint "https: / / example-news-api.com / v2 / latest" to retrieve the latest news data in JSON format.

[1082] News information analysis

[1083] Natural language processing tools

[1084] The server analyzes the collected news data using natural language processing (NLP) techniques. NLP libraries (e.g., spaCy and NLTK) are used to analyze the content of news articles and identify important keywords and sentence structures.

[1085] Example: The server analyzes the news article "Nikkei average stock price rises 2%" and extracts keywords such as "Nikkei average," "stock price," and "2% rise."

[1086] Extraction of important information

[1087] Important information extraction means

[1088] Based on the analysis results, the server filters and extracts important information necessary for investment decisions (for example, causes of stock price fluctuations, corporate earnings announcements, etc.).

[1089] Example: From the news data analyzed by the server, specific information such as "The main reason for the 2% rise in the Nikkei average stock price is the improved performance of export companies" is extracted.

[1090] Generate a summary

[1091] Summary generation means

[1092] Based on the key information extracted by the server, a summarization algorithm is used to generate a summary that is easy for investors to understand. Abstractive summarization and extractive summarization techniques can be combined.

[1093] Example: The server generates a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[1094] Save Summary

[1095] Database storage means

[1096] The server stores the generated summaries in a database, including the summary content, a link to the original news article, and the date and time the summary was generated.

[1097] Example: The server stores in an SQL database the summary "The Nikkei average rose 2%, primarily due to improved performance of export companies" along with further information.

[1098] Provision to users

[1099] User terminal provision means

[1100] When a user accesses the Yahoo Finance portal, the server retrieves the latest summary from the database and provides it to the user's device, allowing the user to make investment decisions based on it.

[1101] Example: When a user visits the "Latest News" section of Yahoo Finance, they are presented with a server-provided summary that reads, "Nikkei average rises 2%, primarily due to improved performance of export companies."

[1102] The system of the present invention includes these means and is capable of collecting, analyzing, summarizing and providing investment-related news quickly and efficiently, thereby greatly assisting investors in obtaining information.

[1103] The processing flow will be explained below.

[1104] Specific processing steps of the news gathering and analysis system

[1105] Step 1:

[1106] The server configures the URL and API endpoint of the news source, for example, registering the endpoint of a particular financial news site or data provider.

[1107] Step 2:

[1108] The server retrieves news data from multiple news sources according to a pre-defined schedule, for example, calling the API every day at 8:00 to retrieve the latest news.

[1109] Step 3:

[1110] The news data retrieved by the server is preprocessed, specifically by extracting text from HTML content and converting it into a parseable format.

[1111] Step 4:

[1112] The server uses natural language processing (NLP) techniques to analyze the preprocessed news data, using NLP libraries to analyze sentence structure and extract keywords.

[1113] Step 5:

[1114] The server then uses the analysis results to extract key information from news articles, filtering the information based on specific keywords such as "stock prices," "earnings announcements," and "market trends."

[1115] Step 6:

[1116] The server organizes the extracted key information and applies a summary generation algorithm to generate summaries that are easy for investors to understand. Abstractive summaries and extractive summaries are used.

[1117] Step 7:

[1118] The server stores the generated summary in a database, including the summary content, the URL of the original article, and the date and time the summary was generated.

[1119] Step 8:

[1120] A user visits the Yahoo! Finance portal and requests to view breaking news.

[1121] Step 9:

[1122] The server retrieves the latest summary data from the database and sends it to the user terminal.

[1123] Step 10:

[1124] The user's device displays the received summary data on the Yahoo Finance interface, allowing the user to view the information.

[1125] Example 1

[1126] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1127] Because news information is vast and scattered, investors do not have time to research and analyze it individually, making it difficult to quickly and efficiently obtain important news information. Furthermore, analyzing news information and generating summaries requires advanced technology and knowledge, and doing this manually is labor-intensive. Therefore, there is a need for a system that can quickly and accurately provide the important news information investors need.

[1128] 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.

[1129] In this invention, the server includes means for periodically collecting news information, means for preprocessing the collected news information, means for analyzing the preprocessed news information using natural language processing, means for extracting important data from the analyzed information, means for generating summaries based on the extracted important data, means for storing the generated summaries in a database, and means for providing the stored summaries to a user terminal, thereby enabling investors to quickly and efficiently obtain the latest important news information and make investment decisions.

[1130] "Means for regularly collecting news information" refers to a function that allows you to set the URL or API endpoint of a specific news source and obtain news data daily or at a fixed time.

[1131] "Means for preprocessing collected news information" refers to a function for removing unnecessary information from acquired news data and converting it into a format that can be analyzed.

[1132] "Means for analyzing preprocessed news information using natural language processing" refers to a function that uses natural language processing techniques and libraries to analyze preprocessed news text and identify important keywords and sentence structures.

[1133] "Means of extracting important data from analyzed information" refers to a function that filters and selects important information necessary for investment decisions from the data obtained as a result of natural language processing.

[1134] The "means for generating a summary based on the extracted important data" is a function that uses an algorithm to create a summary based on the extracted important information in a form that is easy for investors to understand.

[1135] "Means for saving the generated summary in a database" refers to a function that saves the generated summary and its related information in a database so that it can be referenced or searched later.

[1136] The "means for providing the stored summary to the user terminal" is a function for retrieving the latest summary from the database and displaying it on the user terminal.

[1137] The system of the present invention collects, analyzes, summarizes, and provides investment-related news quickly and efficiently. When this system is implemented, the server, terminals, and users work together to perform specific processes. A specific embodiment of this system is described below.

[1138] Gathering news information

[1139] The server configures the URL and API endpoint of a specific news source. News sources include major financial news sites and data providers. For example, the server accesses the API endpoint of a financial news site every day at 8:00 AM to retrieve the latest news data. Specifically, consider the server sending a GET request to the API endpoint "https: / / example-news-api.com / v2 / latest" to retrieve the news data in JSON format.

[1140] Preprocessing of news information

[1141] The news data acquired by the server is preprocessed before analysis. This preprocessing involves extracting only the news text and removing HTML tags and certain unnecessary characters. This converts the data into a format suitable for natural language processing.

[1142] News information analysis

[1143] The server analyzes the preprocessed news data using natural language processing (NLP) technology. SpaCy or NLTK can be used as an NLP library. NLP technology analyzes the content of news articles and identifies important keywords and sentence structures. For example, the server analyzes the news article "Nikkei Stock Average Rise 2%" and extracts keywords such as "Nikkei Stock Average," "Stock Price," and "2% Rise."

[1144] Extraction of important information

[1145] Based on the analysis results, the server filters and extracts important information necessary for investment decisions. Important information includes the causes of stock price fluctuations and corporate earnings announcements. For example, the server extracts specific information such as "The main reason for the 2% rise in the Nikkei average is the improved performance of export companies."

[1146] Generate a summary

[1147] Based on the key information extracted by the server, a summarization algorithm is used to generate summaries that are easy for investors to understand. Abstractive and extractive summarization techniques are used to generate summaries. For example, the server might generate a summary such as, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[1148] Save Summary

[1149] The server saves the generated summary in a database. The data saved in the database includes the summary content, a link to the original news article, the date and time the summary was generated, etc. Specifically, the server saves the summary "The Nikkei average rose 2%, mainly due to improved performance of export companies" in an SQL database.

[1150] Provision to users

[1151] When a user accesses a financial portal site, the server retrieves the latest summary from the database and provides it to the user's device. The user can make investment decisions based on this summary. For example, when a user accesses the "Latest News" section of a financial portal site, a summary provided by the server is displayed: "The Nikkei average rose 2%, mainly due to improved performance of export companies."

[1152] Example prompts for generative AI models

[1153] By inputting the following prompt sentences into the generative AI model, the program's processing can be made easier to understand.

[1154] Example: "Describe a program that retrieves the latest investment-related news from financial news sites, uses natural language processing techniques to extract key points, and generates summaries that are easy for investors to understand."

[1155] This will enable investors to obtain the necessary information quickly and efficiently and make investment decisions.

[1156] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1157] Step 1:

[1158] The server loads the news source configuration. The server reads a previously saved configuration file and obtains the URLs and API endpoints of multiple news sources. The input is the contents of the configuration file, and the output is a list of news sources. Specifically, the server parses the "config / news_sources.json" file and stores the endpoints of each news source in a list.

[1159] Step 2:

[1160] The server periodically collects news data. Using the endpoints of the loaded news sources, the server sends API requests daily or at regular intervals to obtain the latest news data. The input is a list of news sources, and the output is news data in JSON format. Specifically, the server sends a GET request to each API endpoint at 8:00 every day and saves the received data as JSON.

[1161] Step 3:

[1162] The server preprocesses the acquired news data. The server removes unnecessary information from the acquired news data and cleans up the text data. The input is news data in JSON format, and the output is text data suitable for analysis. Specifically, the server parses the JSON data to extract the news text and removes HTML tags and unnecessary characters.

[1163] Step 4:

[1164] The server analyzes the preprocessed news data using natural language processing technology. The server uses an NLP library (for example, spaCy or NLTK) to analyze the content of the news article and identify important keywords and sentence structure. The input is the preprocessed text data, and the output is the analysis results (keywords and sentence structure). Specifically, the server analyzes the news article "Nikkei Stock Average Rise 2%" and extracts keywords such as "Nikkei Stock Average," "Stock Price," and "2% Rise."

[1165] Step 5:

[1166] The server extracts important information from the analysis results. Based on the results of natural language processing, the server filters and extracts important information necessary for investment decisions. The input is the analysis results, and the output is important information. Specifically, the server extracts information such as "The main reason for the 2% rise in the Nikkei average is the improved performance of export companies."

[1167] Step 6:

[1168] The server generates a summary based on the important information. Using a summarization algorithm, the server extracts the important information and creates a summary in a format that is easy for investors to understand. The input is the important information, and the output is the generated summary. Specifically, the server generates the summary, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[1169] Step 7:

[1170] The server saves the generated summary in a database. The server stores the generated summary and its related information in a database. The input is the generated summary and its related information, and the output is the result saved in the database. Specifically, the server saves the summary "Nikkei average rises 2%, mainly due to improved performance of export companies" in an SQL database.

[1171] Step 8:

[1172] When a user accesses the portal site, the server provides the latest summary to the user's device. The server retrieves the latest summary from the database and sends it to the user's device. The input is the user's request, and the output is the summary data. In concrete terms, when a user accesses the "Latest News" section of the portal site, the server displays the summary "Nikkei average rises 2%, mainly due to improved performance of export companies."

[1173] (Application example 1)

[1174] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1175] There is a need to quickly and efficiently collect, analyze, summarize, and provide investors with a vast amount of investment-related news information. However, performing these tasks manually takes time and effort, and there is a risk of missing important timing for investment decisions. In addition, technology is needed to unify information management from different news sources and appropriately extract and summarize important information.

[1176] 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.

[1177] In this invention, the server includes means for periodically collecting news information, means for analyzing the collected news information using natural language processing, and means for extracting important data from the analyzed information, thereby enabling investment-related news to be collected, analyzed, and summarized quickly and efficiently, and provided to investors in a timely manner.

[1178] "News information" refers to current events and information collected from specific news sources.

[1179] "Periodic collection means" refers to a function that automatically obtains news information at specific time intervals.

[1180] "Means of analysis using natural language processing" refers to a technology that converts text data into a form that a computer can understand and analyzes the content.

[1181] "Means for extracting important data" refers to the function of identifying and extracting important information necessary for investment decisions from analyzed news information.

[1182] "Means for generating a summary" refers to a function that concisely summarizes the content based on the extracted important information.

[1183] "Means for storing in a database" refers to the function of storing the generated summaries in an organized manner so that they can be searched and referenced later.

[1184] "Means for providing to user terminal" refers to the function of displaying and transmitting the stored summary information to the user's device.

[1185] "Means for users to access summaries via a web application" means the ability for users to view news summaries on a web platform via the Internet.

[1186] A system for implementing the present invention includes multiple means for quickly and efficiently collecting, analyzing, summarizing, storing, and presenting investment-related news.

[1187] Gathering news information

[1188] The server pre-configures the API endpoint of a specific news source and periodically collects news data. For example, this process can be performed using the Python Requests library. For example, the server periodically accesses the news API to retrieve the latest news data. The collected data is stored in JSON format.

[1189] News information analysis

[1190] The server analyzes the collected news data using natural language processing (NLP) techniques, such as the spaCy library in Python, to identify important keywords and sentence structures from the content of news articles.

[1191] Extraction of important information

[1192] The server then uses the analysis results to extract key data needed to make investment decisions. This is done through keyword matching and rule-based filtering. For example, keywords such as "stock price" and "rise" are extracted from news articles.

[1193] Generate a summary

[1194] The server uses a summarization algorithm to generate a summary based on the extracted important information. This process uses the Python transformers library (e.g., the BERT model). For example, a summary such as "Stock prices rose 2% due to improved performance of export companies" is generated.

[1195] Save Summary

[1196] The server stores the generated summary in an SQL database. The stored data includes the summary content, a link to the original news article, the date and time the summary was generated, etc. For example, a summary such as "Stock prices rose 2% due to improved performance of export companies" is stored in the database.

[1197] Providing a summary

[1198] Users can access news summaries through a web application, which uses the Python Flask framework. For example, when a user accesses the web application, the server retrieves the latest summaries from the database and provides them to the user.

[1199] Hardware and software used

[1200] Hardware: Servers, database servers, web servers

[1201] Software: Python, Flask, spaCy, transformers, sqlite3

[1202] Prompt Sentence Examples

[1203] "Implement a system to collect the latest investment news, analyze it, extract key information, and generate summaries to provide to users."

[1204] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1205] Step 1:

[1206] The server accesses the news source's API endpoint and periodically retrieves news information. The input is the news source's API endpoint, and the output is news data in JSON format. The server uses Python's requests library to send requests to the news API and retrieves the news data received as a response.

[1207] Step 2:

[1208] The server analyzes the acquired news data using natural language processing (NLP) techniques. The input is news data in JSON format, and the output is a list of important keywords contained in the news articles. The server uses the Python spaCy library to extract important keywords and sentence structure from the content of the news articles. Specifically, it analyzes the text of the news articles and performs tokenization and part-of-speech tagging.

[1209] Step 3:

[1210] The server extracts key data based on the analysis results. The input is the analysis results of the news article, and the output is a list of key information needed to make investment decisions. Through keyword matching and rule-based filtering, the server identifies and extracts information containing key keywords such as "stock price," "rise," and "performance."

[1211] Step 4:

[1212] The server uses a summarization algorithm to generate a summary based on the extracted key information. The input is a list of key information, and the output is the summarized text. The server uses the Python transformers library and a generative AI model (e.g., the BERT model) to generate a concise summary of the key information.

[1213] Step 5:

[1214] The server stores the generated summaries in an SQL database. The input is the summarized text, and the output is the records stored in the database. The server uses Python's sqlite3 library to store the summary content, a link to the original news article, and the date and time the summary was generated in the database.

[1215] Step 6:

[1216] A user accesses the latest news summaries through a web application. The input is an access request from the user, and the output is a user interface that displays the latest news summaries. The server uses the Python Flask framework to retrieve the latest summaries from the database and provide them to the user's device. Specifically, when a user sends a request through the web application, the server queries the database, retrieves the latest summaries, and displays them to the user.

[1217] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1218] The system of the present invention not only quickly and efficiently collects, analyzes, summarizes, and provides investment-related news, but also recognizes user emotions and adaptively provides information based on those emotions. To implement this system, the following specific forms are considered and implemented.

[1219] Gathering news information

[1220] News gathering methods

[1221] The server pre-configures the URLs and API endpoints of specific news sources (e.g., major financial news sites or data providers) and periodically collects news data based on these. For example, the server retrieves the latest news from the Wall Street Journal or Reuters APIs at 8:00 every day.

[1222] Example: The server accesses the API endpoint "https: / / example-news-api.com / v2 / latest" to retrieve the latest news data in JSON format.

[1223] News information analysis

[1224] Natural language processing tools

[1225] The server analyzes the collected news data using natural language processing (NLP) techniques. NLP libraries (e.g., spaCy and NLTK) are used to analyze the content of news articles and identify important keywords and sentence structures.

[1226] Example: The server analyzes the news article "Nikkei average stock price rises 2%" and extracts keywords such as "Nikkei average," "stock price," and "2% rise."

[1227] Extraction of important information

[1228] Important information extraction means

[1229] Based on the analysis results, the server filters and extracts important information necessary for investment decisions (for example, causes of stock price fluctuations, corporate earnings announcements, etc.).

[1230] Example: From the news data analyzed by the server, specific information such as "The main reason for the 2% rise in the Nikkei average stock price is the improved performance of export companies" is extracted.

[1231] Generate a summary

[1232] Summary generation means

[1233] Based on the key information extracted by the server, a summarization algorithm is used to generate a summary that is easy for investors to understand. Abstractive summarization and extractive summarization techniques can be combined.

[1234] Example: The server generates a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[1235] Save Summary

[1236] Database storage means

[1237] The server stores the generated summaries in a database, including the summary content, a link to the original news article, and the date and time the summary was generated.

[1238] Example: The server stores in an SQL database the summary "The Nikkei average rose 2%, primarily due to improved performance of export companies" along with further information.

[1239] Provision to users

[1240] User terminal provision means

[1241] When a user accesses the Yahoo Finance portal, the server retrieves the latest summary from the database and provides it to the user's device, allowing the user to make investment decisions based on it.

[1242] Example: When a user visits the "Latest News" section of Yahoo Finance, they are presented with a server-provided summary that reads, "Nikkei average rises 2%, primarily due to improved performance of export companies."

[1243] Introducing the Emotion Engine

[1244] Emotion Engine Means

[1245] An emotion engine is installed on the server to recognize how the user feels about the provided summary information. The emotion engine analyzes the user's operation history, browsing history, click behavior, browsing time, etc. to evaluate the user's emotional state.

[1246] Example: The emotion engine analyzes the user's past behavior history and determines that they tend to have positive emotions toward the news that "the Nikkei average has risen 2%."

[1247] Adaptive Delivery Methods

[1248] Based on the user's emotional state analyzed by the emotion engine, the server adjusts the content and presentation of the summary information provided to the user, for example, by prioritizing the display of content that the user feels positive about.

[1249] Example: Prioritize displaying news about "improved performance of export companies," which users have positive feelings about, along with other related news.

[1250] The system of the present invention includes these means and quickly and efficiently collects, analyzes, summarizes, and provides investment-related news, and also adaptively provides information taking into account the user's emotional state, thereby realizing more personalized investment information provision.

[1251] The processing flow will be explained below.

[1252] Specific processing steps of the news gathering and analysis system (including the sentiment engine)

[1253] Step 1:

[1254] The server configures the URL and API endpoint of the news source, for example, registering the endpoint of a particular financial news site or data provider.

[1255] Step 2:

[1256] The server retrieves news data from multiple news sources according to a pre-defined schedule, for example, calling the API every day at 8:00 to retrieve the latest news.

[1257] Step 3:

[1258] The news data retrieved by the server is preprocessed, specifically by extracting text from HTML content and converting it into a parseable format.

[1259] Step 4:

[1260] The server uses natural language processing (NLP) techniques to analyze the preprocessed news data, using NLP libraries to analyze sentence structure and extract keywords.

[1261] Step 5:

[1262] The server then uses the analysis results to extract key information from news articles, filtering the information based on specific keywords such as "stock prices," "earnings announcements," and "market trends."

[1263] Step 6:

[1264] The server organizes the extracted key information and applies a summary generation algorithm to generate summaries that are easy for investors to understand. Abstractive summaries and extractive summaries are used.

[1265] Step 7:

[1266] The server stores the generated summary in a database, including the summary content, the URL of the original article, and the date and time the summary was generated.

[1267] Step 8:

[1268] A user visits the Yahoo! Finance portal and requests to view breaking news.

[1269] Step 9:

[1270] The server retrieves the latest summary data from the database and sends it to the user terminal.

[1271] Step 10:

[1272] The user's device displays the received summary data on the Yahoo Finance interface, allowing the user to view the information.

[1273] Specific steps for implementing the Emotion Engine

[1274] Step 11:

[1275] The server collects the user's operation history, browsing history, click behavior, browsing time, etc. using an emotion engine.

[1276] Step 12:

[1277] The server analyzes the collected user data and evaluates their emotional state, for example classifying them as positive, negative, or neutral.

[1278] Step 13:

[1279] The server adjusts the content and presentation of the summary information provided based on the user's emotional state, giving priority to content that the user has a positive feeling about.

[1280] Step 14:

[1281] The user views the summary information provided preferentially and accumulates their operation history. The server then analyzes this data using the emotion engine and reflects it in the next information provided.

[1282] Specific examples

[1283] Examples:

[1284] The server retrieves news data such as "Nikkei Stock Average rises 2%" from the financial news API. The server performs NLP analysis to extract keywords such as "Nikkei Stock Average," "stock price," and "2% rise," and applies a summary generation algorithm to generate a summary such as "Nikkei Stock Average rises 2%, primarily due to improved performance of export companies." The server stores this summary in a database, and the user accesses Yahoo! Finance. The server then analyzes the user's past operation history using an emotion engine, and prioritizes displaying news with positive sentiment. This provides users with information they need to make immediate investment decisions.

[1285] Example 2

[1286] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1287] Conventional news information gathering and analysis systems collect, analyze, and summarize news, but they have the problem of being unable to provide information that takes into account the user's emotional state. As a result, it is difficult to provide the information that the user actually needs at the appropriate time, and this can result in excessive or insufficient provision of information.

[1288] 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 means for periodically collecting news data, means for analyzing the collected news data using natural language processing technology, means for extracting important information from the analyzed data, means for generating a summary based on the extracted information, means for saving the generated summary in a storage device, means for providing the saved summary to a user terminal, means for recognizing and evaluating the emotional state of a user, and means for adaptively providing information based on the evaluation result. This makes it possible to provide information adaptively taking the emotional state of a user into consideration.

[1289] "News data" is text data about current events and information, obtained from multiple news sources.

[1290] "Natural language processing technology" is a technology for processing human language using a computer, and involves analyzing text and extracting keywords.

[1291] "Storage" refers to a storage device or database for storing generated summaries.

[1292] A "usage terminal" is a device such as a computer, smartphone, or tablet that a user uses to receive information.

[1293] "User's emotional state" refers to the positive or negative emotional state that a user feels when viewing news information.

[1294] "Adaptive information provision" refers to providing information in a form optimized for a user's emotional state, taking into account that user's emotional state.

[1295] "News source" refers to the source of information that provides news data, including online newspapers, news sites, API services, etc.

[1296] This invention is a system that not only quickly and efficiently collects, analyzes, summarizes, and provides investment-related news, but also recognizes the user's emotional state and adaptively provides information based on that.

[1297] The server periodically collects news data using pre-configured news source URLs or API endpoints. For example, it uses the API endpoints of major financial news sites and data providers. Specifically, the server accesses the Wall Street Journal or Reuters API endpoint "https: / / example-news-api.com / v2 / latest" at 8:00 every day, and retrieves the latest news in JSON format.

[1298] Next, the server analyzes the collected news data using natural language processing (NLP) technology. NLP libraries such as spaCy and NLTK are used to analyze the content of news articles and identify important keywords and sentence structures. For example, the server analyzes the news article "Nikkei Stock Average Rise 2%" and extracts the keywords "Nikkei Stock Average," "Stock Price," and "Rise 2%."

[1299] Next, the server extracts important information based on the analysis results. It filters and extracts information necessary for investment decisions (e.g., causes of stock price fluctuations, corporate earnings announcements, etc.). For example, the server might extract specific information such as, "The main reason the Nikkei average rose 2% is the improved performance of export companies."

[1300] The server then uses a summarization algorithm based on the key information extracted to generate a summary that is easy for investors to understand. Abstractive and extractive summarization techniques can be combined. For example, the server generates a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[1301] The server stores the generated summary in a storage device, such as an SQL database. The stored data includes the summary content, a link to the original news article, the date and time the summary was generated, etc. For example, the server stores the summary "The Nikkei average rose 2%, mainly due to improved performance of export companies" and its detailed information in the SQL database.

[1302] When a user accesses the Yahoo! Finance portal, the server retrieves the latest summary from the database and provides it to the user's device. The user can then make investment decisions based on this summary. For example, when a user accesses the "Latest News" section of Yahoo! Finance, the server provides a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[1303] Furthermore, the server is equipped with an emotion engine that recognizes and evaluates the user's emotional state. The emotion engine analyzes the user's operation history, browsing history, click behavior, browsing time, etc. to evaluate the user's emotional state. For example, the emotion engine analyzes the user's past operation history and determines that the user tends to have positive emotions toward the news that "the Nikkei average rose 2%."

[1304] Based on the user's emotional state analyzed by the emotion engine, the server adjusts the content and display method of the summary information provided to the user's device. It sets the display to prioritize content that the user has a positive feeling about. For example, news about "improving performance of export companies," which the user has a positive feeling about, is displayed with priority along with other related news.

[1305] Prompt Sentence Examples

[1306] "Collect the latest financial news from the Wall Street Journal and extract and summarize the key information."

[1307] "Analyze this news article and identify key keywords."

[1308] "If a user has a positive sentiment towards a particular news item, prioritize showing other related news items."

[1309] In this way, the system of the present invention performs all processes from collecting news data to providing information to the user in one place, and realizes adaptive information provision based on the emotional state of the user.

[1310] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1311] Step 1: Gathering news data

[1312] The server loads the URL or API endpoint of a pre-configured news source, sends a request to the news source's API at the specified time, and retrieves the latest news. Specifically, the server sends a GET request to the API endpoint (e.g., "https: / / example-news-api.com / v2 / latest") at 8:00 every day, and receives the latest news data in JSON format.

[1313] Input: News source URL and API endpoint

[1314] Output: News data in JSON format

[1315] Step 2: Analyzing the news data

[1316] The server analyzes the collected news data using natural language processing technology. An NLP library (e.g., spaCy or NLTK) is used to analyze the content of news articles and identify sentence structure and important keywords. For example, the server analyzes a news article titled "Nikkei Stock Average Rise 2%" and extracts the keywords "Nikkei Stock Average," "Stock Price," and "2% Rise."

[1317] Input: News data in JSON format

[1318] Output: Parsed text data and keywords

[1319] Step 3: Extracting important information

[1320] Based on the analysis results, the server filters and extracts important information necessary for investment decisions. Specifically, the server extracts the specific information that "the main reason for the 2% rise in the Nikkei average is the improved performance of export companies."

[1321] Input: Parsed text data and keywords

[1322] Output: Extracted important information

[1323] Step 4: Generate a summary

[1324] The server uses a summarization algorithm to generate an easy-to-understand summary based on the key information extracted. For example, the server may generate a summary such as, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[1325] Input: Extracted sensitive information

[1326] Output: Generated summary

[1327] Step 5: Save the summary

[1328] The server saves the generated summary in a database. Specifically, the database stores the summary content, a link to the original news article, the date and time of generation, etc. For example, the server saves the summary "Nikkei average rises 2%, mainly due to improved performance of export companies" along with the link and the date and time of generation in an SQL database.

[1329] Input: Generated summary

[1330] Output: Summary and detailed information stored in a database

[1331] Step 6: Provide to users

[1332] When a user accesses a specified portal, the server retrieves the latest summary from the database and provides it to the user's device. For example, when a user accesses the "Latest News" section, the server displays a summary that reads, "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[1333] Input: User request and summary stored in database

[1334] Output: Latest summary displayed on user terminal

[1335] Step 7: Assess your emotional state

[1336] The server uses an emotion engine to analyze the user's operation history, browsing history, click behavior, browsing time, etc., and evaluates the user's emotional state. Specifically, the emotion engine analyzes the user's past operation data and determines that the user tends to have positive emotions toward the news that "the Nikkei average rose 2%."

[1337] Input: User operation history, browsing history, click behavior, browsing time, etc.

[1338] Output: Evaluated user emotional state

[1339] Step 8: Adaptive information provision

[1340] Based on the evaluation results of the emotion engine, the server adjusts the content and display method of the summary information provided to the user's device. For example, it prioritizes the display of news about "improving performance of export companies," which the user has positive feelings about.

[1341] Input: Evaluated emotional state of the user and summaries stored in a database

[1342] Output: Related news displayed in priority order according to the user's emotional state

[1343] (Application example 2)

[1344] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1345] Conventional investment news delivery systems are limited to analyzing collected news information and generating summaries, and do not provide information based on the user's individual emotions or interests. This results in users being exposed to a large amount of unnecessary information, making it difficult to quickly obtain the information necessary for investment decisions. Furthermore, the lack of a mechanism for adaptively providing information based on the user's emotional state makes it difficult to provide personalized investment information.

[1346] The specification processing by the specification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for periodically collecting news information, means for analyzing the collected news information using natural language processing, means for extracting important data from the analyzed information, means for generating a summary based on the extracted information, means for saving the generated summary in a database, means for providing the saved summary to a user terminal, means for recognizing the user's emotional state, and means for adaptively adjusting the information to be provided based on the user's emotional state. This enables users to quickly and efficiently obtain personalized investment information based on their own emotions and interests.

[1347] "Means for regularly collecting news information" refers to a system for obtaining news data from specific news sources at regular intervals.

[1348] "Means for analyzing collected news information using natural language processing" refers to a system that analyzes collected news data using natural language processing technology to understand the content and extract important keywords.

[1349] "Means for extracting important data from analyzed information" refers to a system that filters and extracts information and keywords that are important for investment decisions from news data analyzed using natural language processing.

[1350] A "means for generating a summary based on extracted information" is an algorithm or program that uses key data to summarize information in a way that is easy for users to understand.

[1351] The "means for storing the generated summary in a database" refers to a mechanism for storing the generated summary information in a database so that it can be accessed and searched later.

[1352] The "means for providing the stored summary to the user terminal" is a mechanism for presenting the summary information stored in the database to the user terminal.

[1353] The "means for recognizing the user's emotional state" is a mechanism that analyzes the user's operation history, browsing behavior, etc. to evaluate their current emotions and interests.

[1354] The "means for adaptively adjusting the information provided based on the emotional state" is a mechanism that dynamically changes the information provided and the way it is displayed based on the user's emotional evaluation results.

[1355] The system of the present invention quickly and efficiently collects, analyzes, and summarizes investment-related news, and adaptively provides information based on the user's sentiment. Specific means and processes for implementing this system are described below.

[1356] Gathering news information

[1357] News gathering methods

[1358] The server periodically collects news data from specific news sources, such as major financial news sites and data providers, using HTTP client software such as the requests library in Python.

[1359] News information analysis

[1360] Natural language processing tools

[1361] The server analyzes the collected news data using natural language processing (NLP) technology. NLP uses libraries such as spaCy and NLTK to analyze the content of news articles and identify important keywords and sentence structures. This process converts the content of the news data into an understandable format.

[1362] Extraction of important information

[1363] Important information extraction means

[1364] From the analyzed news data, particularly important information is extracted. This can be done using machine learning algorithms or rule-based analysis methods. For example, the causes of stock price fluctuations and corporate earnings announcements that are important for investment decisions are targeted.

[1365] Generate a summary

[1366] Summary generation means

[1367] The server generates a summary based on the extracted important information using a summary generation algorithm. The algorithm can combine abstractive and extractive summarization techniques. Generative AI models such as GPT-3 are particularly effective.

[1368] Save Summary

[1369] Database storage means

[1370] The generated summaries are then stored in a database, which also contains metadata such as the summary content, a link to the original news article, and the date and time the summary was generated. This process is often performed using a cloud database service such as Amazon RDS.

[1371] Provision to users

[1372] User terminal provision means

[1373] When a user accesses the system through a device (such as a smartphone or PC), the server retrieves the latest summary from the database and provides it to the user, allowing the user to obtain important investment information in a timely manner.

[1374] Introducing the Emotion Engine

[1375] Emotion Engine Means

[1376] An emotion engine is installed on the server to evaluate how users feel about the summary information provided. The emotion engine collects and analyzes data such as user operation history, browsing history, click behavior, and browsing time. Machine learning libraries such as TensorFlow and Keras are used for emotion analysis.

[1377] Adaptive Delivery Methods

[1378] Based on the user's emotional state analyzed by the emotion engine, the server adjusts the content and presentation of the summary information it provides. For example, it may prioritize news stories to which the user has responded positively. This dynamic adjustment is an important feature for providing information optimized for each individual user.

[1379] Specific examples

[1380] 1. News gathering: Access the API endpoint "https: / / example-news-api.com / v2 / latest" to get the latest news data.

[1381] 2. News analysis: Analyze the collected news item "Nikkei average stock price rises 2%" and extract keywords such as "Nikkei average," "stock price," and "2% rise."

[1382] 3. Extraction of important information: Extract important information such as "The main reason for the 2% rise in the Nikkei average stock price is the improved performance of export companies."

[1383] 4. Summary generation: Using GPT-3, generate a summary such as "The Nikkei average rose 2%, primarily due to improved performance of export companies."

[1384] 5. Emotion engine: Analyzes the user's past operation history and determines that they tend to have positive emotions toward the news that "Nikkei average rises 2%," and prioritizes displaying content that evokes positive emotions.

[1385] Prompt Sentence Examples

[1386] Prompt for news summary generation:

[1387] News article: "The Nikkei Stock Average rose 2%. This was mainly due to improved performance of export companies."

[1388] Generated summary: "Nikkei average rises 2%, primarily due to improved performance of export companies"

[1389] In this way, the invention can be specifically implemented.

[1390] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1391] Step 1: Get news information

[1392] The server periodically accesses the news API endpoint to retrieve the latest news data. The input is the API endpoint URL, and the output is the retrieved news data (in JSON format). Specifically, it sends an HTTP request using the Python requests library and receives the news data as a response.

[1393] Step 2: Analyzing the news information

[1394] The server analyzes the news data acquired in step 1 using natural language processing (NLP) technology. The input is the acquired news data, and the output is the analyzed important keywords and phrases. Specifically, it uses NLP libraries such as spaCy and NLTK to perform text analysis.

[1395] Step 3: Extracting important information

[1396] The server extracts important information necessary for investment decisions from the data analyzed in step 2. The input is the analyzed keywords and phrases, and the output is the extracted important information. Specifically, it uses machine learning algorithms and rule-based filtering methods to select important information.

[1397] Step 4: Generate a summary

[1398] The server generates a summary using a summary generation algorithm based on the information extracted in step 3. The input is the extracted key information, and the output is the generated summary. Specifically, a generative AI model such as GPT-3 is used to create the summary. An example of a prompt sentence for the generative AI model is, "News article: The Nikkei average stock price rose 2%. This is mainly due to the improved performance of export companies. Generated summary: The Nikkei average rose 2%, mainly due to the improved performance of export companies."

[1399] Step 5: Save the summary

[1400] The server stores the summaries generated in step 4 in a database. The input is the generated summary information, and the output is the summary data stored in the database. Specifically, the server uses an SQL query to insert the summary information into a cloud database such as Amazon RDS.

[1401] Step 6: Deliver to users

[1402] When a user accesses the system from a terminal, the server retrieves the latest summary from the database and provides it to the user's terminal. The input is the user's access request, and the output is the displayed summary information. Specifically, it receives the user's request, retrieves the summary information from the database using an SQL query, and returns it as an HTTP response.

[1403] Step 7: Recognizing your emotional state

[1404] The server analyzes the user's operation history and browsing behavior to evaluate their emotional state. The input is the user's past operation data, and the output is the evaluation result of the user's emotional state. Specifically, emotion analysis is performed using machine learning libraries such as TensorFlow and Keras.

[1405] Step 8: Adaptively adjusting information provision

[1406] The server adjusts the content and presentation of the information it provides based on the emotion evaluation results obtained in step 7. The input is the evaluation result of the emotional state, and the output is the adjusted summary information. Specifically, it reorganizes the list of high-priority information and applies an algorithm to display it in a form optimized for the user.

[1407] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1408] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1409] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1410] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1411] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1412] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1413] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1414] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1415] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1416] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1417] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1418] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1419] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1420] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1421] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1422] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1423] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1424] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1425] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1426] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1427] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1428] The following is further disclosed regarding the above embodiment.

[1429] (Claim 1)

[1430] A means of regularly gathering news information;

[1431] A means for analyzing the collected news information using natural language processing;

[1432] A means for extracting important data from the analyzed information;

[1433] a means for generating a summary based on the extracted information;

[1434] a means for storing the generated summaries in a database;

[1435] means for providing the stored summary to a user terminal;

[1436] A system including:

[1437] (Claim 2)

[1438] 2. The system of claim 1, wherein the means for collecting news information obtains data from a plurality of news sources.

[1439] (Claim 3)

[1440] 2. The system according to claim 1, wherein the means for analyzing news information by natural language processing generates a summary using a text summarization algorithm.

[1441] "Example 1"

[1442] (Claim 1)

[1443] A means of regularly gathering news information;

[1444] a means for preprocessing the collected news information;

[1445] means for analyzing the preprocessed news information using natural language processing;

[1446] A means for extracting important data from the analyzed information;

[1447] A means for generating summaries based on the extracted key data;

[1448] a means for storing the generated summaries in a database;

[1449] means for providing the stored summary to a user terminal;

[1450] A system including:

[1451] (Claim 2)

[1452] 10. The system according to claim 1, wherein news information is collected from a plurality of news sources on a regular basis.

[1453] (Claim 3)

[1454] 2. The system according to claim 1, wherein the means for analyzing news information by natural language processing uses a natural language processing library to identify keywords and sentence structures and extract important data.

[1455] "Application Example 1"

[1456] (Claim 1)

[1457] A means of regularly gathering news information;

[1458] A means for analyzing the collected news information using natural language processing;

[1459] A means for extracting important data from the analyzed information;

[1460] a means for generating a summary based on the extracted information;

[1461] a means for storing the generated summaries in a database;

[1462] means for providing the stored summary to a user terminal;

[1463] a means for users to access the abstract via a web application;

[1464] A system including:

[1465] (Claim 2)

[1466] 2. The system of claim 1, wherein the means for collecting news information obtains data from a plurality of news sources.

[1467] (Claim 3)

[1468] 2. The system according to claim 1, wherein the means for analyzing news information by natural language processing generates a summary using a text summarization algorithm.

[1469] "Example 2: Combining Emotion Engines"

[1470] (Claim 1)

[1471] a means of collecting news data on a regular basis;

[1472] A means of analyzing the collected news data using natural language processing technology;

[1473] means for extracting significant information from the analyzed data;

[1474] means for generating a summary based on the extracted information;

[1475] means for storing the generated summary in a storage device;

[1476] means for providing the stored summary to a user terminal;

[1477] means for recognizing and assessing the emotional state of a user;

[1478] means for adaptively providing information based on the evaluation results;

[1479] A system including:

[1480] (Claim 2)

[1481] 10. The system of claim 1, wherein the means for collecting news data obtains data from a plurality of news sources.

[1482] (Claim 3)

[1483] 2. The system according to claim 1, wherein the means for analyzing news data using natural language processing techniques generates summaries using a text summarization algorithm.

[1484] "Application example 2 when combining emotion engines"

[1485] (Claim 1)

[1486] A means of regularly gathering news information;

[1487] A means for analyzing the collected news information using natural language processing;

[1488] A means for extracting important data from the analyzed information;

[1489] a means for generating a summary based on the extracted information;

[1490] a means for storing the generated summaries in a database;

[1491] means for providing the stored summary to a user terminal;

[1492] a means for recognizing the emotional state of a user;

[1493] means for adaptively adjusting the information provided based on the emotional state;

[1494] A system including:

[1495] (Claim 2)

[1496] 2. The system of claim 1, wherein the means for collecting news information obtains data from a plurality of news sources.

[1497] (Claim 3)

[1498] 2. The system according to claim 1, wherein the means for analyzing news information by natural language processing generates a summary using a text summarization algorithm. [Explanation of symbols]

[1499] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of regularly gathering news information; A means for analyzing the collected news information using natural language processing; A means for extracting important data from the analyzed information; a means for generating a summary based on the extracted information; a means for storing the generated summaries in a database; means for providing the stored summary to a user terminal; A system including:

2. 2. The system of claim 1, wherein the means for gathering news information obtains data from a plurality of news sources.

3. 2. The system according to claim 1, wherein the means for analyzing news information by natural language processing generates summaries using a text summarization algorithm.

Images