System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting data of a financial market; means for multi-modal analysis of the collected data; means for generating a trading strategy based on the analysis; and means for providing the generated trading strategy to an investor.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Modern financial markets are highly complex and dynamic, requiring individual investors, asset managers, hedge funds, and market analysts to analyze massive amounts of market data quickly and efficiently. Early detection of market trends and risks and optimizing the accuracy and timing of investment strategies are extremely challenging. For this reason, traditional analytical methods are often insufficient. This invention aims to address these challenges by providing an advanced market analysis system that utilizes generative AI and AGI to comprehensively analyze multimodal data and detect market trends, risk factors, and investment opportunities in real time. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems by providing a system that includes a means for collecting financial market data, a means for multimodal analysis of the collected data, a means for generating a trading strategy based on the analysis results, and a means for providing the generated trading strategy to investors. Specifically, the system collects at least one of text data, audio data, image data, and video data, and detects market trends and risk factors in real time through multimodal analysis. The generated trading strategy then includes buy or sell recommendations based on the collected and analyzed data, thereby supporting investor decision-making.

[0006] "Financial market data" refers to any digital information related to financial markets, such as stock prices, exchange rates, commodity prices, economic indicators, and news articles.

[0007] "Collection Means" refers to the combination of hardware and software used to obtain the required information from the Data Sources.

[0008] "Multimodal analysis" refers to the process of comprehensively analyzing different types of data, such as text data, audio data, image data, and video data.

[0009] A "trading strategy" refers to a specific set of guidelines or rules for buying and selling assets in financial markets.

[0010] An "investor" is an individual or legal entity that invests money in stocks, bonds, real estate, or other financial products with the aim of earning a profit.

[0011] "System" refers to an integrated hardware and software setup for collecting, analyzing, generating and delivering financial market data.

[0012] "Analysis results" refers to useful information and insights obtained through the data analysis process.

[0013] "Real-time" refers to data processing and response with very short latency.

[0014] A "buy or sell recommendation" means providing guidance on whether to buy or sell a particular financial instrument. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] 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).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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."

[0036] This invention relates to an "Intelligent Market Sensor" system that supports investment analysis in financial markets. This system utilizes generative AI and AGI to comprehensively analyze multimodal data such as text, audio, images, and video to detect market trends, risk factors, and investment opportunities in real time, and provides investors with specific trading strategies.

[0037] Explanation of program processing

[0038] Processing data collection

[0039] The server collects the latest data from various data sources, such as financial news, audio, and images. Specifically, it retrieves the latest news articles from the financial news API, downloads audio data, and retrieves and loads image data. This collected data is then used for subsequent analysis steps.

[0040] Data analysis process

[0041] The server analyzes the collected text, audio, and image data. For text data, it performs sentiment analysis using a generative AI model (e.g., DistilBERT). For audio data, it uses a generative AI model (e.g., Wav2Vec2) to convert audio to text, and then further analyzes the text for sentiment. For image data, it analyzes it using an image classification model (e.g., Vision Transformer) to obtain the information indicated by the image as a label.

[0042] Processing of Trading Strategy Generation

[0043] The server automatically generates a trading strategy based on the analysis results. Specifically, it integrates the results of text, voice, and image analysis, and generates a strategy to "buy" if the analysis results are positive, and "sell" if they are negative. This generated strategy becomes important information to support investors' decision-making.

[0044] Processing of investor offerings

[0045] The server then sends the generated trading strategy to the user's device. The user then accesses the trading platform and executes specific investment actions based on the strategy provided. In this process, the user is able to grasp market trends and risk factors in real time, enabling them to make timely and accurate investment decisions.

[0046] Specific examples

[0047] Consider the following example as a specific scenario.

[0048] 1. Data collection

[0049] For example, the server retrieves text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest, while simultaneously retrieving audio of a financial analyst from https: / / api.financial-audio.com / latest and image data showing rising stock prices from https: / / api.financial-images.com / latest.

[0050] 2. Data Analysis

[0051] For this collected text data, the server uses a generative AI model to obtain the emotion label "positive." After converting the voice data into text using the generative AI model, the server performs a similar emotion analysis and obtains the "positive" label. Furthermore, the generative AI model is used to analyze image data, obtaining the label "stock price rise."

[0052] 3. Generating a Trading Strategy

[0053] The server aggregates all the analysis results and generates a "buy" strategy since the overall result is positive. This strategy provides useful guidance to investors.

[0054] 4. Offering to investors

[0055] The server sends a message to the user's terminal saying, "The current recommended trading strategy is 'BUY'." Based on this information, the user takes specific investment action through the trading platform.

[0056] This "Intelligent Market Sensor" system comprehensively analyzes information from a variety of data sources and generates and provides highly accurate trading strategies in real time, thereby significantly supporting investors' investment decisions.

[0057] The processing flow will be explained below.

[0058] Step 1:

[0059] The server retrieves the latest text data from the financial news API, specifically, collects JSON data of news articles from https: / / api.financial-news.com / latest.

[0060] Step 2:

[0061] The server collects audio data from the financial information service by downloading the audio file from https: / / api.financial-audio.com / latest and saving the data.

[0062] Step 3:

[0063] The server retrieves the latest financial image data from the image API by accessing the endpoint https: / / api.financial-images.com / latest and loading the retrieved image data as an image object using Image.open .

[0064] Step 4:

[0065] The server analyzes the collected text data using a sentiment analysis model. Specifically, it runs a text sentiment analysis pipeline using the DistilBERT model and outputs sentiment labels (positive, neutral, or negative) for news articles.

[0066] Step 5:

[0067] The server uses a model to convert audio data into text. Specifically, it transcribes the audio data using the Wav2Vec2 model and stores the resulting text data.

[0068] Step 6:

[0069] The server analyzes the text extracted from the audio data using a sentiment analysis model. Specifically, it uses the DistilBERT model to obtain sentiment labels (positive, neutral, negative) for the previously saved text data.

[0070] Step 7:

[0071] The server uses an image classification model to analyze the image data. Specifically, it uses a Vision Transformer (ViT) model to obtain labels for the information the image represents (e.g., a graph of rising stock prices).

[0072] Step 8:

[0073] The server integrates all data analysis results, specifically, compiles emotion and image labels obtained from each data source into a single list and performs a comprehensive evaluation.

[0074] Step 9:

[0075] The server automatically generates a trading strategy based on the integrated analysis results. Specifically, if the proportion of positive sentiment labels is high, it generates a "BUY" strategy, and if not, it generates a "SELL" strategy.

[0076] Step 10:

[0077] The server sends the generated trading strategy to the user's terminal. The user can take specific investment actions based on the provided strategy. For example, a message such as "The current recommended trading strategy is 'BUY'" is displayed on the terminal.

[0078] Example 1

[0079] 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."

[0080] In financial markets, it is extremely important for investors to grasp market trends, risk factors, and investment opportunities in real time and make accurate and prompt decisions. However, conventional systems have fragmented information collection, analysis, and strategy generation, making it difficult to comprehensively analyze information from diverse data sources. In particular, there is a need for unified processing of data in different formats, such as text, audio, and images, and for generating and providing reliable trading strategies based on the analysis results.

[0081] 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.

[0082] In this invention, the server includes a means for collecting financial market information, a means for multimodal analysis of the collected information, a means for analyzing text, voice, and image data using a generative AI model, a means for generating a trading strategy based on the analysis results, and a means for providing the generated trading strategy to investors, thereby enabling unified processing of data in different formats and providing reliable strategies in real time.

[0083] "Financial market information" refers to all data necessary to understand market trends, such as stock prices, exchange rates, economic indicators, and corporate financial information.

[0084] "Collect" refers to the act of obtaining data from designated data sources and storing it in the system.

[0085] "Multimodal analysis" refers to the simultaneous analysis of multiple forms of data, such as text, audio, and images, to obtain comprehensive information.

[0086] A "generative AI model" refers to an algorithm or system that uses artificial intelligence techniques to analyze and process data.

[0087] A "trading strategy" refers to an investor's plan or policy for buying and selling in financial markets.

[0088] "Providing to investors" means sending the generated information and strategies to the user's device or platform so that they can be used by investors.

[0089] "Sentiment analysis" refers to a technique for identifying emotions and opinions from text data and quantitatively evaluating them.

[0090] "Buy or sell recommendation" refers to a suggestion based on the results of an analysis as to whether to buy or sell a particular financial asset.

[0091] This invention relates to an "Intelligent Market Sensor" system for supporting investment analysis in financial markets. This system utilizes generative AI and AGI to comprehensively analyze multimodal data such as text, audio, images, and video to detect market trends, risk factors, and investment opportunities in real time, and provides specific trading strategies to investors.

[0092] The server collects the latest data from various data sources, such as financial news, audio, and images. For example, the server retrieves the latest news articles from a financial news API, downloads audio data, and retrieves and loads image data. This collected data is then used for subsequent analysis steps. Generative AI models (e.g., DistilBERT, Wav2Vec2, Vision Transformer) are used to analyze the collected text, audio, and image data.

[0093] First, the server performs sentiment analysis on the collected text data using a generative AI model (e.g., DistilBERT). For audio data, it converts it into text using a generative AI model (e.g., Wav2Vec2), and then further analyzes the text for sentiment analysis. For image data, it analyzes it using an image classification model (e.g., Vision Transformer), and obtains the information indicated by the image as a label.

[0094] The server then automatically generates a trading strategy based on the analysis results. Specifically, it integrates the results of text, voice, and image analysis, and generates a "buy" strategy if the analysis results are positive, and a "sell" strategy if the analysis results are negative. This generated strategy becomes important information to support investors' decision-making.

[0095] Finally, the server sends the generated trading strategy to the user's device. The user then accesses the trading platform based on the strategy and executes specific investment actions. In this process, the user is able to grasp market trends and risk factors in real time, enabling them to make timely and accurate investment decisions.

[0096] Consider the following example as a specific scenario.

[0097] 1. Data collection

[0098] For example, the server retrieves text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest. At the same time, it retrieves audio of a financial analyst from https: / / api.financial-audio.com / latest and image data showing rising stock prices from https: / / api.financial-images.com / latest.

[0099] 2. Data Analysis

[0100] The server uses a generative AI model to obtain the emotion label "positive" from the collected text data. It also converts voice data into text using the generative AI model, then performs emotion analysis and obtains the "positive" label. It also analyzes image data using the generative AI model and obtains the label "stock price rise."

[0101] 3. Generating a Trading Strategy

[0102] The server aggregates all the analysis results and generates a "buy" strategy since the overall result is positive. This strategy provides useful guidance to investors.

[0103] 4. Offering to investors

[0104] The server sends a message to the user's terminal saying, "The current recommended trading strategy is 'BUY'." Based on this information, the user takes specific investment action through the trading platform.

[0105] Here are some example prompts you can enter into your generative AI model:

[0106] 1. Prompts for text sentiment analysis

[0107] "Analyze the sentiment of the following financial news article: The market is currently trending upward."

[0108] 2. Speech-to-text prompts

[0109] "Transcribe this financial analyst's speech to text: Audio data stream"

[0110] 3. Prompt for image analysis

[0111] "Please classify the content of this image and label it appropriately: Image Data Stream"

[0112] In this way, the present invention comprehensively analyzes information from various data sources such as text, audio, and image data, and generates and provides highly accurate trading strategies in real time, thereby greatly supporting investors' investment decisions.

[0113] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0114] Step 1:

[0115] The server retrieves the latest news articles from the financial news API.

[0116] Specifically, the server sends a GET request to the URL https: / / api.financial-news.com / latest and receives news data in text format. The retrieved news articles are stored in JSON format.

[0117] Input: Financial News API URL

[0118] Data processing: GET request, saving received data in JSON format

[0119] Output: News data in JSON format

[0120] Step 2:

[0121] The server downloads the audio data.

[0122] Specifically, the server downloads audio data from https: / / api.financial-audio.com / latest, saves it in WAV format, and later converts it to text.

[0123] Input: URL of audio data

[0124] Data processing: HTTP requests, saving received data in WAV format

[0125] Output: WAV format audio data

[0126] Step 3:

[0127] The server acquires and reads the image data.

[0128] Specifically, the server downloads image data from https: / / api.financial-images.com / latest, saves it in JPEG format, and later feeds it into the image analysis model.

[0129] Input: Image data URL

[0130] Data processing: HTTP requests, saving received data in JPEG format

[0131] Output: JPEG format image data

[0132] Step 4:

[0133] The server performs sentiment analysis on the collected text data using a generative AI model.

[0134] Specifically, the server inputs text data into a generative AI model (e.g., DistilBERT) to obtain sentiment labels. This step uses the prompt sentence, "Analyze the sentiment of the following financial news article: The market is currently trending upward."

[0135] Input: JSON format text data, prompt statement

[0136] Data Processing: Sentiment Analysis with Generative AI Models

[0137] Output: Sentiment label (e.g. "positive")

[0138] Step 5:

[0139] The server converts the audio data into text using a generative AI model.

[0140] Specifically, the server inputs WAV-formatted audio data into a generative AI model (e.g., Wav2Vec2) and obtains text data. This procedure uses the prompt "Convert this financial analyst's speech into text: Audio data stream."

[0141] Input: WAV format audio data, prompt text

[0142] Data Processing: Speech-to-Text Conversion with Generative AI Models

[0143] Output: Text data

[0144] Step 6:

[0145] The server performs sentiment analysis on the text data converted from the voice data.

[0146] Specifically, the server inputs the converted text data into a generative AI model (e.g., DistilBERT) to obtain a sentiment label. This step uses the prompt sentence, "Analyze the sentiment of the following text: The market is trending upward."

[0147] Input: Text data, prompt

[0148] Data Processing: Sentiment Analysis with Generative AI Models

[0149] Output: Sentiment label (e.g. "positive")

[0150] Step 7:

[0151] The server analyzes the image data with an image classification model.

[0152] Specifically, the server inputs image data into a generative AI model (e.g., Vision Transformer) to obtain image labels. This procedure uses the prompt "Please classify the content of this image and provide an appropriate label: image data stream."

[0153] Input: JPEG image data, prompt text

[0154] Data Processing: Image Classification with Generative AI Models

[0155] Output: Image label (e.g. "Stock price rise")

[0156] Step 8:

[0157] The server integrates the results of text, voice and image analysis.

[0158] Specifically, the server integrates the results of each analysis and determines an overall emotion label. For example, if all emotion labels are "positive," the overall result is also determined to be positive.

[0159] Input: Text analysis results, voice analysis results, image analysis results

[0160] Data processing: Integrating each result and determining the overall sentiment

[0161] Output: Overall sentiment label (e.g. "positive")

[0162] Step 9:

[0163] The server generates a trading strategy based on the analysis results.

[0164] Specifically, the server automatically generates a strategy based on the integrated analysis results. For example, if the analysis results are positive, it generates a strategy to "buy," and if they are negative, it generates a strategy to "sell."

[0165] Input: Overall emotion label

[0166] Data processing: Automatic generation of strategies

[0167] Output: Trading strategy (e.g. "Buy")

[0168] Step 10:

[0169] The server transmits the generated trading strategy to the user's terminal.

[0170] As a specific operation, the server transmits the generated strategy to the user's terminal in the form of a message.

[0171] Input: Trading Strategy

[0172] Data processing: Message generation and transmission

[0173] Output: A message to the user terminal (e.g., "The current recommended trading strategy is 'BUY'")

[0174] Step 11:

[0175] Based on the provided strategy, the user accesses the trading platform and executes specific investment actions.

[0176] Specifically, the user operates the trading platform based on the message received from the server and executes the instructed investment action. For example, if a "BUY" recommendation is received, the user will purchase the corresponding stock.

[0177] Input: Trading strategy message from the server

[0178] Data processing: input to trading platform

[0179] Output: Specific investment action (e.g., purchase of stock)

[0180] (Application example 1)

[0181] 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."

[0182] Conventional financial market analysis systems rely on limited information sources for data collection and analysis, and lack data diversification to improve the accuracy of trading strategies provided to investors. On the other hand, systems that use user activity data to recommend individually optimized content also face the challenge of making appropriate recommendations based on real-time data analysis. Therefore, the present invention aims to provide highly accurate trading strategies and individually optimized content by collecting and analyzing financial market data and user activity data in a multimodal manner.

[0183] 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.

[0184] In this invention, the server includes means for collecting financial market data, means for multimodal analysis of the collected data, means for generating a trading strategy based on the analysis results, means for providing the generated trading strategy to investors, means for collecting user activity data, means for analyzing the collected user activity data to generate optimal content, and means for providing the generated content to users. This makes it possible to comprehensively analyze data related to financial markets and user activity and provide appropriate investment strategies and content in real time.

[0185] "Financial market data" refers to information related to market trends, such as stock prices, exchange rates, interest rates, and financial reports.

[0186] "Multimodal analysis" refers to a method of comprehensive analysis using multiple data formats such as text, audio, images, and video.

[0187] A "trading strategy" refers to a specific plan for determining the timing and method of buying and selling in financial markets.

[0188] "Means for providing to investors" refers to methods and techniques for communicating the generated trading strategies to investors.

[0189] "User activity data" refers to information related to a user's behavior, such as a user's browsing history, ratings, and feedback.

[0190] "Means for generating optimal content" refers to methods and technologies for recommending and creating content that is most suitable for users based on collected user activity data.

[0191] "Means for providing generated content to a user" refers to methods and technologies for providing content suitable for a user to a user's device or application.

[0192] This invention relates to a system that collects and analyzes financial market data and user activity data in a multimodal manner, and provides highly accurate trading strategies and individually optimized content.

[0193] System Configuration

[0194] 1. Hardware

[0195] Server: Responsible for data collection, analysis, trading strategy and content generation. Equipped with a high-performance processor and large memory capacity.

[0196] User devices: Smartphones, smart glasses, head-mounted displays, etc. Providing trading strategies and content to users.

[0197] 2. Software

[0198] Prediction models: DistilBERT is used for text analysis, Wav2Vec2 for speech analysis, and Vision Transformer for image analysis.

[0199] API: RESTful API for collecting financial news, audio data, image data, user activity data, etc.

[0200] Database: A database for storing collected data and analysis results.

[0201] Program processing

[0202] Data collection

[0203] The server collects financial market data and user activity data, for example, using the following APIs:

[0204] Get the latest market information from our financial news API.

[0205] Collect trending data from social media APIs.

[0206] Obtain user browsing history from the content platform API.

[0207] Data analysis

[0208] The server analyzes the collected data. Specific examples include:

[0209] The text data is subjected to sentiment analysis using the DistilBERT model to determine whether it is positive or negative.

[0210] The audio data is converted into text using the Wav2Vec2 model, and then sentiment analysis is performed.

[0211] Image data is analyzed using the Vision Transformer model, and the information indicated by the image is obtained as labels.

[0212] Trading Strategies and Content Generation

[0213] The server generates trading strategies and content based on the analysis results.

[0214] If the analysis result is positive, a trading strategy of "buy" is generated, and if it is negative, a trading strategy of "sell" is generated.

[0215] Recommend the most suitable content based on user activity data.

[0216] Provision to users

[0217] The generated trading strategies and content are sent to the user's terminal, where the user can make investment decisions and view the content based on the information provided.

[0218] It sends notifications to smartphones and smart glasses apps, displaying current trading strategy recommendations and content to users.

[0219] Specific examples

[0220] For example, the server retrieves text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest, and simultaneously collects user browsing history data from https: / / api.content-platform.com / user_activity.

[0221] The server then uses a generative AI model to generate a "positive" sentiment label for the collected text data. It then performs a similar analysis on the user's browsing history data to identify patterns of content that the user prefers.

[0222] Finally, the server integrates all the analysis results and generates a "buy" strategy since the overall results are positive. It also recommends content that the user will like and sends a notification to the user's smartphone, such as "The current recommended content is 'Latest Movie Trends'."

[0223] Prompt Sentence Examples

[0224] "Analyze users' browsing history and feedback over the past six months to recommend content that is likely to generate a positive response."

[0225] As described above, the present invention is a system that can comprehensively analyze financial market and user activity data and provide two important functions, investment strategies and content recommendations, in real time.

[0226] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0227] Step 1:

[0228] The server collects financial market data and user activity data. Specifically, it obtains the latest market information from a financial news API, collects trend-related data from a social media API, and obtains user browsing history from a content platform API. In this case, the input is financial market information and user activity data, and the output is the collected data.

[0229] Step 2:

[0230] The server analyzes the collected data. Specifically, it performs sentiment analysis on the collected text data using the DistilBERT model, converts the audio data to text using the Wav2Vec2 model, then performs sentiment analysis, and analyzes the image data using the Vision Transformer model to obtain the information indicated by the image as a label. The inputs are text data, audio data, and image data, and the output is the analysis results: emotion labels and image labels.

[0231] Step 3:

[0232] The server generates trading strategies and content based on the analysis results. For example, if the analysis results of text, audio, and images are positive, a trading strategy of "buy" is generated, and if they are negative, a trading strategy of "sell" is generated. The server also recommends optimal content based on user activity data. In this case, the input is the analysis results and user activity data, and the output is the generated trading strategy and content.

[0233] Step 4:

[0234] The server provides the generated trading strategies and content to the user's device. The server notifies the user's smartphone or smart glasses app of the recommended trading strategies and content, allowing the user to access them at any time. In this case, the input is the generated trading strategies and content, and the output is the information provided to the user.

[0235] Step 5:

[0236] Based on the provided trading strategies and content, users take appropriate investment actions and view content. Specifically, they execute trades on the trading platform based on notifications or view recommended content. In this case, the input is the generated information and the output is the user's actions.

[0237] 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.

[0238] This invention relates to an "intelligent market sensor" system that supports investment analysis in financial markets. This system utilizes generative AI and AGI to comprehensively analyze multimodal data, including text, voice, images, and video, to detect market trends, risk factors, and investment opportunities in real time and provide specific trading strategies to investors. Furthermore, this system incorporates an emotion engine that recognizes user emotions, enabling it to optimize trading strategies taking user emotions into account.

[0239] Explanation of program processing

[0240] Processing data collection

[0241] The server acquires text data from the financial news API, collects audio data from the financial information service, and acquires image data from the image API, and subjects these data to an analysis step.

[0242] Processing emotion data collection

[0243] The server collects emotion-related data from the user's text messages, voice inputs, and biometric sensors, including the user's heart rate and facial expression analysis data.

[0244] Data analysis process

[0245] The server analyzes the collected text data using a generative AI model to obtain emotion labels. It also analyzes voice data, converts it into text, and obtains emotion labels. It analyzes image data using an image classification model to classify the image content as a label.

[0246] Processing of sentiment data analysis

[0247] The server uses an emotion engine to analyze data obtained from the user's text, voice, and biometric sensors to recognize the user's emotional state, which is then used in the subsequent trading strategy generation step.

[0248] Processing of Trading Strategy Generation

[0249] The server automatically generates a trading strategy based on the analysis results and the user's emotion data. Specifically, if the emotion label of the market data is positive and the user's emotion is stable, the server generates a "buy" strategy, and if it is negative, the server generates a "sell" strategy.

[0250] Processing of investor offerings

[0251] The server then sends the generated trading strategy to the user's device and notifies the user. The user then accesses the trading platform based on the strategy provided and executes specific investment actions. In this process, the user is able to grasp market trends and risk factors in real time, enabling them to make timely and accurate investment decisions.

[0252] Specific examples

[0253] Consider the following example as a specific scenario.

[0254] 1. Data collection

[0255] For example, the server retrieves text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest, while simultaneously retrieving audio of a financial analyst from https: / / api.financial-audio.com / latest and image data showing rising stock prices from https: / / api.financial-images.com / latest.

[0256] 2. Collecting Emotional Data

[0257] The server collects the user's input text, speech, and data from biometric sensors (e.g., heart rate data). For example, if a user types, "What do you think about the market these days?", the server collects that text and speech.

[0258] 3. Data Analysis

[0259] For this collected text data, the server uses a generative AI model to obtain the emotion label "positive." After converting the voice data into text using the generative AI model, the server performs a similar emotion analysis and obtains the "positive" label. Furthermore, the generative AI model is used to analyze image data, obtaining the label "stock price rise."

[0260] 4. Emotion Data Analysis

[0261] The server uses an emotion engine to analyze the user's text, voice, and biometric sensor data, recognizing, for example, if the user indicates an emotion such as "excited."

[0262] 5. Generating a Trading Strategy

[0263] The server integrates all the analysis results and the user's emotional state, and generates a "buy" strategy since the overall results are positive and the user's emotions are stable.

[0264] 6. Offering to investors

[0265] The server sends a message to the user's terminal saying, "The current recommended trading strategy is 'BUY'." Based on this information, the user takes specific investment action through the trading platform.

[0266] This system comprehensively analyzes market data and user sentiment data, and generates and provides highly accurate trading strategies in real time, thereby significantly supporting investors' investment decisions.

[0267] The processing flow will be explained below.

[0268] Step 1:

[0269] The server retrieves the latest text data from the financial news API, specifically, collects JSON data of news articles from https: / / api.financial-news.com / latest.

[0270] Step 2:

[0271] The server collects audio data from the financial information service by downloading the audio file from https: / / api.financial-audio.com / latest and saving it.

[0272] Step 3:

[0273] The server retrieves financial image data from the image API by accessing the endpoint https: / / api.financial-images.com / latest and loading the retrieved image data as an image object using Image.open .

[0274] Step 4:

[0275] The server collects the user's input text, specifically, collects messages entered by the user in a chat window of the trading platform as a log.

[0276] Step 5:

[0277] The server collects voice input data from the user, specifically, stores voice data acquired through the voice assistant function.

[0278] Step 6:

[0279] The server collects the user's biometric sensor data, specifically recording heart rate, skin temperature, and other physiological data obtained from biometric sensors such as a smartwatch.

[0280] Step 7:

[0281] The server analyzes the collected text data using a sentiment analysis model. Specifically, it runs a text sentiment analysis pipeline using the DistilBERT model and outputs sentiment labels (positive, neutral, or negative) for news articles.

[0282] Step 8:

[0283] The server uses a model to convert audio data into text. Specifically, it transcribes the audio data using the Wav2Vec2 model and stores the resulting text data.

[0284] Step 9:

[0285] The server analyzes the text extracted from the audio data using a sentiment analysis model. Specifically, it uses the DistilBERT model to obtain sentiment labels (positive, neutral, negative) for the previously saved text data.

[0286] Step 10:

[0287] The server uses an image classification model to analyze the image data. Specifically, it uses a Vision Transformer (ViT) model to obtain labels for the information the image represents (e.g., a graph of rising stock prices).

[0288] Step 11:

[0289] The server uses an emotion engine to analyze the user's text, voice, and biometric sensor data to estimate the user's emotional state and obtain emotion labels.

[0290] Step 12:

[0291] The server integrates all data analysis results, specifically, compiles emotion and image labels obtained from each data source into a single list and performs a comprehensive evaluation.

[0292] Step 13:

[0293] The server automatically generates a trading strategy based on the integrated analysis results and user sentiment data. Specifically, if the sentiment analysis results and user sentiment are stable, it generates a "BUY" strategy, and if not, it generates a "SELL" strategy.

[0294] Step 14:

[0295] The server sends the generated trading strategy to the user's terminal, notifying the user by displaying a message saying, "The current recommended trading strategy is 'BUY'."

[0296] Step 15:

[0297] The user accesses the trading platform based on the provided trading strategy and takes specific investment actions, such as "buying" or "selling" according to the displayed strategy.

[0298] In this way, the server can comprehensively analyze market data and user sentiment data, and generate and provide highly accurate trading strategies in real time, thereby significantly supporting users' investment decisions.

[0299] Example 2

[0300] 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."

[0301] It is extremely difficult for investors to make instant decisions amidst the rapid fluctuations in information and vast amounts of data in financial markets. Furthermore, because investors' emotional state has a significant impact on investment decisions, strategies that ignore emotions have limited effectiveness. Therefore, in order for investors to make faster and more accurate decisions, a system is needed that can comprehensively analyze market data and emotional data and provide investors with specific trading strategies in real time.

[0302] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0303] In this invention, the server includes means for collecting financial market data, means for performing multimodal analysis of the collected data and user emotion data, means for generating a trading strategy based on the analysis results and the user emotion data, and means for providing the generated trading strategy to investors, thereby enabling investors to grasp market trends and obtain highly accurate trading strategies that take their own emotional state into account in real time.

[0304] "Financial market data" refers to various information obtained from financial markets, including stock prices, exchange rates, interest rates, economic indicators, news articles, reports, etc.

[0305] "Multimodal analysis" is a method for integrating and analyzing data of different formats (text, audio, images, biometric sensors, etc.).

[0306] A "trading strategy" refers to decisions and plans for buying and selling in financial markets, and specifically includes recommendations for "buying" and "selling."

[0307] "Emotion data" is data that represents the psychological state of the user, and includes information obtained from text messages, voice input, biometric sensors, and the like.

[0308] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to analyze data and generate a specific outcome (e.g., a sentiment label or a trading strategy).

[0309] A "prompt sentence" is an input sentence that causes a generative AI model to perform a specific analysis or generation task.

[0310] This invention relates to a system for supporting investment analysis in financial markets, specifically an "Intelligent Market Sensor" system. This system uses generative AI models and AGI to comprehensively analyze multimodal data, including text, voice, images, and biometric sensors, to detect market trends, risk factors, and investment opportunities in real time, and provide specific trading strategies to investors.

[0311] Data collection

[0312] The server retrieves text data from a financial news API and collects audio data from a financial information service. It also retrieves image data from an image API and subjects these data to an analysis step. For example, the server may retrieve text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest. At the same time, it retrieves audio of a financial analyst's speech from https: / / api.financial-audio.com / latest and image data showing rising stock prices from https: / / api.financial-images.com / latest.

[0313] Emotional Data Collection

[0314] The server collects emotion-related data from the user's text messages, speech input, and biometric sensors. This data includes, for example, the user's text input (e.g., "What do you think about the market these days?"), speech, and biometric data such as heart rate.

[0315] Data analysis

[0316] The server analyzes the collected text data using a generative AI model (for example, OpenAI's GPT-4) to obtain an emotion label. It also analyzes voice data in the same way, converts it into text, and then obtains an emotion label. This conversion is performed using Google's Speech-to-Text API. Image data is analyzed using an image classification model such as Google's Vision API, and the content is classified as a label. For example, text data such as "The market is currently on the rise" is input into a generative AI model to create a prompt that obtains the emotion label "positive."

[0317] Emotional Data Analysis

[0318] The server uses an emotion engine to analyze data obtained from the user's text, voice, and biometric sensors to recognize the user's emotional state. For example, it identifies emotions such as "excited" from the user's heart rate and facial expression data. This recognized emotion data is used in the subsequent trading strategy generation step.

[0319] Trading Strategy Generation

[0320] The server automatically generates a trading strategy based on the analysis results and the user's emotional data. Specifically, if the market data's emotional label is positive and the user's emotional state is stable, a "buy" strategy is generated; if it is negative, a "sell" strategy is generated. The server prompts the generating AI model with the following prompt: "The current market data is positive. The user's emotional state is also stable. What trading strategy do you recommend?"

[0321] Offering to investors

[0322] The server then sends the generated trading strategy to the user's device and notifies the user. Based on the strategy, the user can access the trading platform and take specific investment actions. This process allows users to grasp market trends and risk factors in real time and make timely and accurate investment decisions.

[0323] This system comprehensively analyzes market data and user sentiment data, and generates and provides highly accurate trading strategies in real time, thereby significantly supporting investors' investment decisions.

[0324] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0325] Step 1:

[0326] Data collection

[0327] The server first obtains text data from a financial news API (e.g., https: / / api.financial-news.com / latest). It receives the response data (news article text) from the API as input and stores the text data in an internal database as output. At the same time, it collects audio data from a financial information service and obtains the audio file using an audio API (e.g., https: / / api.financial-audio.com / latest). It also obtains image data from an image API (e.g., https: / / api.financial-images.com / latest) and stores them in an internal database. Specifically, the server sends an HTTPS request and parses the JSON-formatted response to extract the necessary information.

[0328] Step 2:

[0329] Emotional Data Collection

[0330] The server collects emotion-related data from text messages entered by the user, audio from conversations, and biometric sensors. As input, it receives text messages, audio files, and biometric sensor data sent from the user's device. As output, it generates structured data formatted for analysis. For example, if a user types, "What do you think about the market these days?", the server collects the text and also captures the audio data and saves it as an audio file.

[0331] Step 3:

[0332] Data analysis

[0333] The server inputs the collected text data into a generative AI model (e.g., OpenAI's GPT-4) to obtain an emotion label. The collected text data is provided as input, and a generated emotion label (e.g., "positive" or "negative") is obtained as output. Voice data is also converted into text using voice recognition software (e.g., Google's Speech-to-Text API), which is then analyzed by the generative AI model. Image data is analyzed using an image classification model (e.g., Google's Vision API), and the content of the image is classified as a label. Specifically, the analysis results in label information such as "the market is on the rise," "financial analysts' comments are positive," and "image showing rising stock prices."

[0334] Step 4:

[0335] Emotional Data Analysis

[0336] The server uses an emotion engine to analyze data obtained from the user's text, voice, and biometric sensors to recognize the user's emotional state. The server provides the user's text data, voice data, and biometric data as input, and identifies the user's emotional state (e.g., "excited" or "calm") as output. Specifically, the server runs the text analysis engine, voice analysis engine, and biometric sensor analysis engine to evaluate the user's overall emotional state.

[0337] Step 5:

[0338] Trading Strategy Generation

[0339] The server automatically generates trading strategies based on market data and user sentiment data. It provides analysis results and sentiment data as input, and generates specific trading strategies such as "buy" or "sell" as output. This is done by inputting prompts such as "Current market data is positive. The user's sentiment state is stable. What trading strategy do you recommend?" into the generative AI model, and obtaining output from the model. For example, if the overall market data is positive, a "buy" strategy is generated.

[0340] Step 6:

[0341] Offering to investors

[0342] The server sends the generated trading strategy to the user's device and notifies the user. It provides the generated trading strategy data as input and sends a notification in the form of a message to the user's device as output. The user receives the notification and accesses a trading platform (e.g., an online securities site or a dedicated app) to perform specific investment actions. Specifically, the server sends a notification to the user's device via a REST API, and the user can make investments through the trading application.

[0343] (Application example 2)

[0344] 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."

[0345] Conventional trading strategy generation systems are limited to analyzing financial market data and do not address the optimization of sales strategies in brick-and-mortar stores. Furthermore, because strategies are not generated taking into account the emotional state of customers, it is difficult to improve customer satisfaction and sales efficiency. Therefore, there is a need for a method to optimize brick-and-mortar store sales strategies in real time, reflecting the emotional state of customers.

[0346] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0347] In this invention, the server includes means for collecting financial market data and data from a physical store, means for performing multimodal analysis of the collected data and recognizing customer emotion data, means for generating trading and sales strategies based on the analysis results and the customer emotion data, and means for providing the generated trading and sales strategies. This makes it possible to generate sales strategies based on product trends and customer emotions in the physical store in addition to analyzing financial market data, thereby improving customer satisfaction and maximizing sales.

[0348] "Financial market data" refers to information such as text, audio, images, and video related to financial markets, including stock prices, trading volume, corporate financial status, economic indicators, and news.

[0349] "In-store data" refers to information such as customer behavior, purchase history, inventory status, and promotional effectiveness in physical stores, and includes data obtained from IoT sensors and cameras.

[0350] "Multimodal analysis" is a technology that integrates and analyzes multiple data formats, such as text, audio, images, and video, and is a method of extracting useful information from data using generative AI models.

[0351] "Customer emotion data" refers to the emotional state analyzed based on data obtained from the customer's facial expressions, voice, text messages, and biometric sensors, and includes emotional categories such as positive, negative, excited, and calm.

[0352] A "trading strategy" is a plan that outlines the method or direction for executing transactions in financial markets, including buying and selling decisions.

[0353] A "sales strategy" is a plan that outlines the methods and policies for selling products in physical stores, including strengthening promotions and adjusting inventory.

[0354] A "generative AI model" is a model that uses artificial intelligence techniques to analyze collected data and is trained to perform specific tasks, such as pattern recognition and predictions in the data.

[0355] "Emotion engine" refers to technology that recognizes and labels emotions from a user's text, voice, and biometric sensors, and is used to perform data analysis based on their emotional state.

[0356] "Increased promotion" refers to increasing sales promotion activities such as advertising and discounts for specific products or services in order to promote product sales.

[0357] An "IoT sensor" is a sensor connected to the Internet that measures and collects physical environmental data and transmits the data in real time.

[0358] "Sales maximization" refers to management and strategic activities aimed at increasing the profitability of a store or business as much as possible.

[0359] The present invention is a system that collects and analyzes data from financial markets and data from physical stores, and provides users with optimal trading and sales strategies. Specific embodiments for carrying out the present invention will be described below.

[0360] System Configuration

[0361] Hardware and Software

[0362] 1. Server: A central processing unit that collects data, analyzes it, and generates strategies. It has a high-performance processor and sufficient storage.

[0363] 2. IoT Sensors: These sensors collect environmental data within physical stores, measuring temperature, humidity, movement, etc.

[0364] 3. Camera: Captures customer movements and facial expressions in the physical store and sends them to the server as image data.

[0365] 4. Voice Assistant: A device that conducts voice interactions with customers and transmits voice data to a server.

[0366] 5. Generative AI models: Machine learning models that analyze text, voice, and image data to generate investment and sales strategies.

[0367] 6. Emotion engine: Software technology for analyzing the user's emotional state and obtaining emotion labels.

[0368] Data collection methods

[0369] The server collects financial market data and data from physical stores. Financial market data includes text data, audio data, image data, and video data obtained from news APIs and financial information providers. Physical store data is collected from IoT sensors, cameras, and voice assistants.

[0370] Data Analysis Methods

[0371] The server uses generative AI models to perform multimodal analysis of the collected data: text data is analyzed using natural language processing (NLP) techniques, voice data is converted to text and then labeled with an emotion label, and image data is analyzed using image classification models to recognize the customer's emotional state.

[0372] Emotion data analysis method

[0373] The server uses an emotion engine to analyze data from users' text, voice, and biometric sensors to recognize their emotional state. For example, if a customer is expressing positive emotions, it can provide strategies based on that.

[0374] Strategy Generation Method

[0375] Based on the analysis results and customer sentiment data, the server generates trading and sales strategies, for example, if the financial market data is positive and the customer sentiment state is positive, it will suggest a "buy" trading strategy or strengthen the promotion of a specific product.

[0376] Strategy delivery methods

[0377] The server provides the generated trading and sales strategies to the users who will actually use them, and the users then take specific investment and sales actions based on these strategies.

[0378] Specific examples

[0379] For example, the server retrieves market text data from the URL "https: / / api.financial-news.com / latest" and assigns an emotion label of "positive." It also analyzes customer facial expressions through cameras in physical stores to determine whether the customer is enjoying themselves. Based on this data, the server generates a message saying, "The current recommended trading strategy is 'BUY'," and sends it to the user's device. It also suggests to the store operator to strengthen promotions of specific products.

[0380] Example prompt sentence:

[0381] "Analyze recent customer sentiment and correlate it with point-of-sale data to optimize your sales strategy."

[0382] "How can we update the product's availability and increase promotions when we recognize high customer satisfaction for a particular product?"

[0383] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0384] Step 1:

[0385] Data collection

[0386] The server collects financial market data from news APIs, financial information services, image APIs, etc. For example, it obtains text data from "https: / / api.financial-news.com / latest," audio data from "https: / / api.financial-audio.com / latest," and image data from "https: / / api.financial-images.com / latest." It also simultaneously collects data from IoT sensors, cameras, and voice assistants in physical stores. Various APIs are used as input, and the collected text, audio, and image data are obtained as output.

[0387] Step 2:

[0388] Data Preprocessing

[0389] The server converts the collected raw data into a format suitable for analysis: it standardizes text data, converts voice data into text using voice recognition technology, and preprocesses image data to standardize pixel information. It receives the collected text, voice, and image data as input, and outputs a standardized dataset.

[0390] Step 3:

[0391] Emotional Data Collection

[0392] The server collects user input text, voice, and biometric sensor data. Specifically, voice data is obtained through a voice assistant, and biometric sensor data is collected from heart rate and facial expression analysis. The input is text, voice, and biometric sensor data from the user, and the output is data that can be labeled with emotion.

[0393] Step 4:

[0394] Data analysis

[0395] The server uses a generative AI model to analyze collected text, audio, and image data to obtain emotion labels. For example, it performs natural language processing (NLP) on text data, emotion recognition on audio data, and content classification on image data. It receives a standardized dataset as input and obtains emotion labels as output.

[0396] Step 5:

[0397] Emotional Data Analysis

[0398] The server uses an emotion engine to comprehensively analyze the user's emotional data, recognize the user's emotional state from text, voice, and biometric sensor data, and generate an emotion label based on this. It receives the data collected from the user as input and obtains the user's emotional state data as output.

[0399] Step 6:

[0400] Strategy Generation

[0401] The server generates trading and sales strategies based on the analysis results and user emotion data, for example, making buying and selling decisions or proposing promotion enhancements. It receives emotion labels and analysis results as input and obtains specific strategies as output.

[0402] Step 7:

[0403] Strategy Providing

[0404] The server provides the generated trading and selling strategies to the user. Specifically, it sends a message to the user's terminal saying "The current recommended trading strategy is 'BUY'." It receives the generated strategies as input and displays a notification message on the user's terminal as output.

[0405] 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.

[0406] 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.

[0407] 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.

[0408] [Second embodiment]

[0409] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0410] 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.

[0411] 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).

[0412] 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.

[0413] 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.

[0414] 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).

[0415] 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.

[0416] 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.

[0417] 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.

[0418] 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.

[0419] In the smart glasses 214, 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.

[0420] 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."

[0421] This invention relates to an "Intelligent Market Sensor" system that supports investment analysis in financial markets. This system utilizes generative AI and AGI to comprehensively analyze multimodal data such as text, audio, images, and video to detect market trends, risk factors, and investment opportunities in real time, and provides investors with specific trading strategies.

[0422] Explanation of program processing

[0423] Processing data collection

[0424] The server collects the latest data from various data sources, such as financial news, audio, and images. Specifically, it retrieves the latest news articles from the financial news API, downloads audio data, and retrieves and loads image data. This collected data is then used for subsequent analysis steps.

[0425] Data analysis process

[0426] The server analyzes the collected text, audio, and image data. For text data, it performs sentiment analysis using a generative AI model (e.g., DistilBERT). For audio data, it uses a generative AI model (e.g., Wav2Vec2) to convert audio to text, and then further analyzes the text for sentiment. For image data, it analyzes it using an image classification model (e.g., Vision Transformer) to obtain the information indicated by the image as a label.

[0427] Processing of Trading Strategy Generation

[0428] The server automatically generates a trading strategy based on the analysis results. Specifically, it integrates the results of text, voice, and image analysis, and generates a strategy to "buy" if the analysis results are positive, and "sell" if they are negative. This generated strategy becomes important information to support investors' decision-making.

[0429] Processing of investor offerings

[0430] The server then sends the generated trading strategy to the user's device. The user then accesses the trading platform and executes specific investment actions based on the strategy provided. In this process, the user is able to grasp market trends and risk factors in real time, enabling them to make timely and accurate investment decisions.

[0431] Specific examples

[0432] Consider the following example as a specific scenario.

[0433] 1. Data collection

[0434] For example, the server retrieves text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest, while simultaneously retrieving audio of a financial analyst from https: / / api.financial-audio.com / latest and image data showing rising stock prices from https: / / api.financial-images.com / latest.

[0435] 2. Data Analysis

[0436] For this collected text data, the server uses a generative AI model to obtain the emotion label "positive." After converting the voice data into text using the generative AI model, the server performs a similar emotion analysis and obtains the "positive" label. Furthermore, the generative AI model is used to analyze image data, obtaining the label "stock price rise."

[0437] 3. Generating a Trading Strategy

[0438] The server aggregates all the analysis results and generates a "buy" strategy since the overall result is positive. This strategy provides useful guidance to investors.

[0439] 4. Offering to investors

[0440] The server sends a message to the user's terminal saying, "The current recommended trading strategy is 'BUY'." Based on this information, the user takes specific investment action through the trading platform.

[0441] This "Intelligent Market Sensor" system comprehensively analyzes information from a variety of data sources and generates and provides highly accurate trading strategies in real time, thereby significantly supporting investors' investment decisions.

[0442] The processing flow will be explained below.

[0443] Step 1:

[0444] The server retrieves the latest text data from the financial news API, specifically, collects JSON data of news articles from https: / / api.financial-news.com / latest.

[0445] Step 2:

[0446] The server collects audio data from the financial information service by downloading the audio file from https: / / api.financial-audio.com / latest and saving the data.

[0447] Step 3:

[0448] The server retrieves the latest financial image data from the image API by accessing the endpoint https: / / api.financial-images.com / latest and loading the retrieved image data as an image object using Image.open .

[0449] Step 4:

[0450] The server analyzes the collected text data using a sentiment analysis model. Specifically, it runs a text sentiment analysis pipeline using the DistilBERT model and outputs sentiment labels (positive, neutral, or negative) for news articles.

[0451] Step 5:

[0452] The server uses a model to convert audio data into text. Specifically, it transcribes the audio data using the Wav2Vec2 model and stores the resulting text data.

[0453] Step 6:

[0454] The server analyzes the text extracted from the audio data using a sentiment analysis model. Specifically, it uses the DistilBERT model to obtain sentiment labels (positive, neutral, negative) for the previously saved text data.

[0455] Step 7:

[0456] The server uses an image classification model to analyze the image data. Specifically, it uses a Vision Transformer (ViT) model to obtain labels for the information the image represents (e.g., a graph of rising stock prices).

[0457] Step 8:

[0458] The server integrates all data analysis results, specifically, compiles emotion and image labels obtained from each data source into a single list and performs a comprehensive evaluation.

[0459] Step 9:

[0460] The server automatically generates a trading strategy based on the integrated analysis results. Specifically, if the proportion of positive sentiment labels is high, it generates a "BUY" strategy, and if not, it generates a "SELL" strategy.

[0461] Step 10:

[0462] The server sends the generated trading strategy to the user's terminal. The user can take specific investment actions based on the provided strategy. For example, a message such as "The current recommended trading strategy is 'BUY'" is displayed on the terminal.

[0463] Example 1

[0464] 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."

[0465] In financial markets, it is extremely important for investors to grasp market trends, risk factors, and investment opportunities in real time and make accurate and prompt decisions. However, conventional systems have fragmented information collection, analysis, and strategy generation, making it difficult to comprehensively analyze information from diverse data sources. In particular, there is a need for unified processing of data in different formats, such as text, audio, and images, and for generating and providing reliable trading strategies based on the analysis results.

[0466] 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.

[0467] In this invention, the server includes a means for collecting financial market information, a means for multimodal analysis of the collected information, a means for analyzing text, voice, and image data using a generative AI model, a means for generating a trading strategy based on the analysis results, and a means for providing the generated trading strategy to investors, thereby enabling unified processing of data in different formats and providing reliable strategies in real time.

[0468] "Financial market information" refers to all data necessary to understand market trends, such as stock prices, exchange rates, economic indicators, and corporate financial information.

[0469] "Collect" refers to the act of obtaining data from designated data sources and storing it in the system.

[0470] "Multimodal analysis" refers to the simultaneous analysis of multiple forms of data, such as text, audio, and images, to obtain comprehensive information.

[0471] A "generative AI model" refers to an algorithm or system that uses artificial intelligence techniques to analyze and process data.

[0472] A "trading strategy" refers to an investor's plan or policy for buying and selling in financial markets.

[0473] "Providing to investors" means sending the generated information and strategies to the user's device or platform so that they can be used by investors.

[0474] "Sentiment analysis" refers to a technique for identifying emotions and opinions from text data and quantitatively evaluating them.

[0475] "Buy or sell recommendation" refers to a suggestion based on the results of an analysis as to whether to buy or sell a particular financial asset.

[0476] This invention relates to an "Intelligent Market Sensor" system for supporting investment analysis in financial markets. This system utilizes generative AI and AGI to comprehensively analyze multimodal data such as text, audio, images, and video to detect market trends, risk factors, and investment opportunities in real time, and provides specific trading strategies to investors.

[0477] The server collects the latest data from various data sources, such as financial news, audio, and images. For example, the server retrieves the latest news articles from a financial news API, downloads audio data, and retrieves and loads image data. This collected data is then used for subsequent analysis steps. Generative AI models (e.g., DistilBERT, Wav2Vec2, Vision Transformer) are used to analyze the collected text, audio, and image data.

[0478] First, the server performs sentiment analysis on the collected text data using a generative AI model (e.g., DistilBERT). For audio data, it converts it into text using a generative AI model (e.g., Wav2Vec2), and then further analyzes the text for sentiment analysis. For image data, it analyzes it using an image classification model (e.g., Vision Transformer), and obtains the information indicated by the image as a label.

[0479] The server then automatically generates a trading strategy based on the analysis results. Specifically, it integrates the results of text, voice, and image analysis, and generates a "buy" strategy if the analysis results are positive, and a "sell" strategy if the analysis results are negative. This generated strategy becomes important information to support investors' decision-making.

[0480] Finally, the server sends the generated trading strategy to the user's device. The user then accesses the trading platform based on the strategy and executes specific investment actions. In this process, the user is able to grasp market trends and risk factors in real time, enabling them to make timely and accurate investment decisions.

[0481] Consider the following example as a specific scenario.

[0482] 1. Data collection

[0483] For example, the server retrieves text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest. At the same time, it retrieves audio of a financial analyst from https: / / api.financial-audio.com / latest and image data showing rising stock prices from https: / / api.financial-images.com / latest.

[0484] 2. Data Analysis

[0485] The server uses a generative AI model to obtain the emotion label "positive" from the collected text data. It also converts voice data into text using the generative AI model, then performs emotion analysis and obtains the "positive" label. It also analyzes image data using the generative AI model and obtains the label "stock price rise."

[0486] 3. Generating a Trading Strategy

[0487] The server aggregates all the analysis results and generates a "buy" strategy since the overall result is positive. This strategy provides useful guidance to investors.

[0488] 4. Offering to investors

[0489] The server sends a message to the user's terminal saying, "The current recommended trading strategy is 'BUY'." Based on this information, the user takes specific investment action through the trading platform.

[0490] Here are some example prompts you can enter into your generative AI model:

[0491] 1. Prompts for text sentiment analysis

[0492] "Analyze the sentiment of the following financial news article: The market is currently trending upward."

[0493] 2. Speech-to-text prompts

[0494] "Transcribe this financial analyst's speech to text: Audio data stream"

[0495] 3. Prompt for image analysis

[0496] "Please classify the content of this image and label it appropriately: Image Data Stream"

[0497] In this way, the present invention comprehensively analyzes information from various data sources such as text, audio, and image data, and generates and provides highly accurate trading strategies in real time, thereby greatly supporting investors' investment decisions.

[0498] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0499] Step 1:

[0500] The server retrieves the latest news articles from the financial news API.

[0501] Specifically, the server sends a GET request to the URL https: / / api.financial-news.com / latest and receives news data in text format. The retrieved news articles are stored in JSON format.

[0502] Input: Financial News API URL

[0503] Data processing: GET request, saving received data in JSON format

[0504] Output: News data in JSON format

[0505] Step 2:

[0506] The server downloads the audio data.

[0507] Specifically, the server downloads audio data from https: / / api.financial-audio.com / latest, saves it in WAV format, and later converts it to text.

[0508] Input: URL of audio data

[0509] Data processing: HTTP requests, saving received data in WAV format

[0510] Output: WAV format audio data

[0511] Step 3:

[0512] The server acquires and reads the image data.

[0513] Specifically, the server downloads image data from https: / / api.financial-images.com / latest, saves it in JPEG format, and later feeds it into the image analysis model.

[0514] Input: Image data URL

[0515] Data processing: HTTP requests, saving received data in JPEG format

[0516] Output: JPEG format image data

[0517] Step 4:

[0518] The server performs sentiment analysis on the collected text data using a generative AI model.

[0519] Specifically, the server inputs text data into a generative AI model (e.g., DistilBERT) to obtain sentiment labels. This step uses the prompt sentence, "Analyze the sentiment of the following financial news article: The market is currently trending upward."

[0520] Input: JSON format text data, prompt statement

[0521] Data Processing: Sentiment Analysis with Generative AI Models

[0522] Output: Sentiment label (e.g. "positive")

[0523] Step 5:

[0524] The server converts the audio data into text using a generative AI model.

[0525] Specifically, the server inputs WAV-formatted audio data into a generative AI model (e.g., Wav2Vec2) and obtains text data. This procedure uses the prompt "Convert this financial analyst's speech into text: Audio data stream."

[0526] Input: WAV format audio data, prompt text

[0527] Data Processing: Speech-to-Text Conversion with Generative AI Models

[0528] Output: Text data

[0529] Step 6:

[0530] The server performs sentiment analysis on the text data converted from the voice data.

[0531] Specifically, the server inputs the converted text data into a generative AI model (e.g., DistilBERT) to obtain a sentiment label. This step uses the prompt sentence, "Analyze the sentiment of the following text: The market is trending upward."

[0532] Input: Text data, prompt

[0533] Data Processing: Sentiment Analysis with Generative AI Models

[0534] Output: Sentiment label (e.g. "positive")

[0535] Step 7:

[0536] The server analyzes the image data with an image classification model.

[0537] Specifically, the server inputs image data into a generative AI model (e.g., Vision Transformer) to obtain image labels. This procedure uses the prompt "Please classify the content of this image and provide an appropriate label: image data stream."

[0538] Input: JPEG image data, prompt text

[0539] Data Processing: Image Classification with Generative AI Models

[0540] Output: Image label (e.g. "Stock price rise")

[0541] Step 8:

[0542] The server integrates the results of text, voice and image analysis.

[0543] Specifically, the server integrates the results of each analysis and determines an overall emotion label. For example, if all emotion labels are "positive," the overall result is also determined to be positive.

[0544] Input: Text analysis results, voice analysis results, image analysis results

[0545] Data processing: Integrating each result and determining the overall sentiment

[0546] Output: Overall sentiment label (e.g. "positive")

[0547] Step 9:

[0548] The server generates a trading strategy based on the analysis results.

[0549] Specifically, the server automatically generates a strategy based on the integrated analysis results. For example, if the analysis results are positive, it generates a strategy to "buy," and if they are negative, it generates a strategy to "sell."

[0550] Input: Overall emotion label

[0551] Data processing: Automatic generation of strategies

[0552] Output: Trading strategy (e.g. "Buy")

[0553] Step 10:

[0554] The server transmits the generated trading strategy to the user's terminal.

[0555] As a specific operation, the server transmits the generated strategy to the user's terminal in the form of a message.

[0556] Input: Trading Strategy

[0557] Data processing: Message generation and transmission

[0558] Output: A message to the user terminal (e.g., "The current recommended trading strategy is 'BUY'")

[0559] Step 11:

[0560] Based on the provided strategy, the user accesses the trading platform and executes specific investment actions.

[0561] Specifically, the user operates the trading platform based on the message received from the server and executes the instructed investment action. For example, if a "BUY" recommendation is received, the user will purchase the corresponding stock.

[0562] Input: Trading strategy message from the server

[0563] Data processing: input to trading platform

[0564] Output: Specific investment action (e.g., purchase of stock)

[0565] (Application example 1)

[0566] 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."

[0567] Conventional financial market analysis systems rely on limited information sources for data collection and analysis, and lack data diversification to improve the accuracy of trading strategies provided to investors. On the other hand, systems that use user activity data to recommend individually optimized content also face the challenge of making appropriate recommendations based on real-time data analysis. Therefore, the present invention aims to provide highly accurate trading strategies and individually optimized content by collecting and analyzing financial market data and user activity data in a multimodal manner.

[0568] 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.

[0569] In this invention, the server includes means for collecting financial market data, means for multimodal analysis of the collected data, means for generating a trading strategy based on the analysis results, means for providing the generated trading strategy to investors, means for collecting user activity data, means for analyzing the collected user activity data to generate optimal content, and means for providing the generated content to users. This makes it possible to comprehensively analyze data related to financial markets and user activity and provide appropriate investment strategies and content in real time.

[0570] "Financial market data" refers to information related to market trends, such as stock prices, exchange rates, interest rates, and financial reports.

[0571] "Multimodal analysis" refers to a method of comprehensive analysis using multiple data formats such as text, audio, images, and video.

[0572] A "trading strategy" refers to a specific plan for determining the timing and method of buying and selling in financial markets.

[0573] "Means for providing to investors" refers to methods and techniques for communicating the generated trading strategies to investors.

[0574] "User activity data" refers to information related to a user's behavior, such as a user's browsing history, ratings, and feedback.

[0575] "Means for generating optimal content" refers to methods and technologies for recommending and creating content that is most suitable for users based on collected user activity data.

[0576] "Means for providing generated content to a user" refers to methods and technologies for providing content suitable for a user to a user's device or application.

[0577] This invention relates to a system that collects and analyzes financial market data and user activity data in a multimodal manner, and provides highly accurate trading strategies and individually optimized content.

[0578] System Configuration

[0579] 1. Hardware

[0580] Server: Responsible for data collection, analysis, trading strategy and content generation. Equipped with a high-performance processor and large memory capacity.

[0581] User devices: Smartphones, smart glasses, head-mounted displays, etc. Providing trading strategies and content to users.

[0582] 2. Software

[0583] Prediction models: DistilBERT is used for text analysis, Wav2Vec2 for speech analysis, and Vision Transformer for image analysis.

[0584] API: RESTful API for collecting financial news, audio data, image data, user activity data, etc.

[0585] Database: A database for storing collected data and analysis results.

[0586] Program processing

[0587] Data collection

[0588] The server collects financial market data and user activity data, for example, using the following APIs:

[0589] Get the latest market information from our financial news API.

[0590] Collect trending data from social media APIs.

[0591] Obtain user browsing history from the content platform API.

[0592] Data analysis

[0593] The server analyzes the collected data. Specific examples include:

[0594] The text data is subjected to sentiment analysis using the DistilBERT model to determine whether it is positive or negative.

[0595] The audio data is converted into text using the Wav2Vec2 model, and then sentiment analysis is performed.

[0596] Image data is analyzed using the Vision Transformer model, and the information indicated by the image is obtained as labels.

[0597] Trading Strategies and Content Generation

[0598] The server generates trading strategies and content based on the analysis results.

[0599] If the analysis result is positive, a trading strategy of "buy" is generated, and if it is negative, a trading strategy of "sell" is generated.

[0600] Recommend the most suitable content based on user activity data.

[0601] Provision to users

[0602] The generated trading strategies and content are sent to the user's terminal, where the user can make investment decisions and view the content based on the information provided.

[0603] It sends notifications to smartphones and smart glasses apps, displaying current trading strategy recommendations and content to users.

[0604] Specific examples

[0605] For example, the server retrieves text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest, and simultaneously collects user browsing history data from https: / / api.content-platform.com / user_activity.

[0606] The server then uses a generative AI model to generate a "positive" sentiment label for the collected text data. It then performs a similar analysis on the user's browsing history data to identify patterns of content that the user prefers.

[0607] Finally, the server integrates all the analysis results and generates a "buy" strategy since the overall results are positive. It also recommends content that the user will like and sends a notification to the user's smartphone, such as "The current recommended content is 'Latest Movie Trends'."

[0608] Prompt Sentence Examples

[0609] "Analyze users' browsing history and feedback over the past six months to recommend content that is likely to generate a positive response."

[0610] As described above, the present invention is a system that can comprehensively analyze financial market and user activity data and provide two important functions, investment strategies and content recommendations, in real time.

[0611] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0612] Step 1:

[0613] The server collects financial market data and user activity data. Specifically, it obtains the latest market information from a financial news API, collects trend-related data from a social media API, and obtains user browsing history from a content platform API. In this case, the input is financial market information and user activity data, and the output is the collected data.

[0614] Step 2:

[0615] The server analyzes the collected data. Specifically, it performs sentiment analysis on the collected text data using the DistilBERT model, converts the audio data to text using the Wav2Vec2 model, then performs sentiment analysis, and analyzes the image data using the Vision Transformer model to obtain the information indicated by the image as a label. The inputs are text data, audio data, and image data, and the output is the analysis results: emotion labels and image labels.

[0616] Step 3:

[0617] The server generates trading strategies and content based on the analysis results. For example, if the analysis results of text, audio, and images are positive, a trading strategy of "buy" is generated, and if they are negative, a trading strategy of "sell" is generated. The server also recommends optimal content based on user activity data. In this case, the input is the analysis results and user activity data, and the output is the generated trading strategy and content.

[0618] Step 4:

[0619] The server provides the generated trading strategies and content to the user's device. The server notifies the user's smartphone or smart glasses app of the recommended trading strategies and content, allowing the user to access them at any time. In this case, the input is the generated trading strategies and content, and the output is the information provided to the user.

[0620] Step 5:

[0621] Based on the provided trading strategies and content, users take appropriate investment actions and view content. Specifically, they execute trades on the trading platform based on notifications or view recommended content. In this case, the input is the generated information and the output is the user's actions.

[0622] 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.

[0623] This invention relates to an "intelligent market sensor" system that supports investment analysis in financial markets. This system utilizes generative AI and AGI to comprehensively analyze multimodal data, including text, voice, images, and video, to detect market trends, risk factors, and investment opportunities in real time and provide specific trading strategies to investors. Furthermore, this system incorporates an emotion engine that recognizes user emotions, enabling it to optimize trading strategies taking user emotions into account.

[0624] Explanation of program processing

[0625] Processing data collection

[0626] The server acquires text data from the financial news API, collects audio data from the financial information service, and acquires image data from the image API, and subjects these data to an analysis step.

[0627] Processing emotion data collection

[0628] The server collects emotion-related data from the user's text messages, voice inputs, and biometric sensors, including the user's heart rate and facial expression analysis data.

[0629] Data analysis process

[0630] The server analyzes the collected text data using a generative AI model to obtain emotion labels. It also analyzes voice data, converts it into text, and obtains emotion labels. It analyzes image data using an image classification model to classify the image content as a label.

[0631] Processing of sentiment data analysis

[0632] The server uses an emotion engine to analyze data obtained from the user's text, voice, and biometric sensors to recognize the user's emotional state, which is then used in the subsequent trading strategy generation step.

[0633] Processing of Trading Strategy Generation

[0634] The server automatically generates a trading strategy based on the analysis results and the user's emotion data. Specifically, if the emotion label of the market data is positive and the user's emotion is stable, the server generates a "buy" strategy, and if it is negative, the server generates a "sell" strategy.

[0635] Processing of investor offerings

[0636] The server then sends the generated trading strategy to the user's device and notifies the user. The user then accesses the trading platform based on the strategy provided and executes specific investment actions. In this process, the user is able to grasp market trends and risk factors in real time, enabling them to make timely and accurate investment decisions.

[0637] Specific examples

[0638] Consider the following example as a specific scenario.

[0639] 1. Data collection

[0640] For example, the server retrieves text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest, while simultaneously retrieving audio of a financial analyst from https: / / api.financial-audio.com / latest and image data showing rising stock prices from https: / / api.financial-images.com / latest.

[0641] 2. Collecting Emotional Data

[0642] The server collects the user's input text, speech, and data from biometric sensors (e.g., heart rate data). For example, if a user types, "What do you think about the market these days?", the server collects that text and speech.

[0643] 3. Data Analysis

[0644] For this collected text data, the server uses a generative AI model to obtain the emotion label "positive." After converting the voice data into text using the generative AI model, the server performs a similar emotion analysis and obtains the "positive" label. Furthermore, the generative AI model is used to analyze image data, obtaining the label "stock price rise."

[0645] 4. Emotion Data Analysis

[0646] The server uses an emotion engine to analyze the user's text, voice, and biometric sensor data, recognizing, for example, if the user indicates an emotion such as "excited."

[0647] 5. Generating a Trading Strategy

[0648] The server integrates all the analysis results and the user's emotional state, and generates a "buy" strategy since the overall results are positive and the user's emotions are stable.

[0649] 6. Offering to investors

[0650] The server sends a message to the user's terminal saying, "The current recommended trading strategy is 'BUY'." Based on this information, the user takes specific investment action through the trading platform.

[0651] This system comprehensively analyzes market data and user sentiment data, and generates and provides highly accurate trading strategies in real time, thereby significantly supporting investors' investment decisions.

[0652] The processing flow will be explained below.

[0653] Step 1:

[0654] The server retrieves the latest text data from the financial news API, specifically, collects JSON data of news articles from https: / / api.financial-news.com / latest.

[0655] Step 2:

[0656] The server collects audio data from the financial information service by downloading the audio file from https: / / api.financial-audio.com / latest and saving it.

[0657] Step 3:

[0658] The server retrieves financial image data from the image API by accessing the endpoint https: / / api.financial-images.com / latest and loading the retrieved image data as an image object using Image.open .

[0659] Step 4:

[0660] The server collects the user's input text, specifically, collects messages entered by the user in a chat window of the trading platform as a log.

[0661] Step 5:

[0662] The server collects voice input data from the user, specifically, stores voice data acquired through the voice assistant function.

[0663] Step 6:

[0664] The server collects the user's biometric sensor data, specifically recording heart rate, skin temperature, and other physiological data obtained from biometric sensors such as a smartwatch.

[0665] Step 7:

[0666] The server analyzes the collected text data using a sentiment analysis model. Specifically, it runs a text sentiment analysis pipeline using the DistilBERT model and outputs sentiment labels (positive, neutral, or negative) for news articles.

[0667] Step 8:

[0668] The server uses a model to convert audio data into text. Specifically, it transcribes the audio data using the Wav2Vec2 model and stores the resulting text data.

[0669] Step 9:

[0670] The server analyzes the text extracted from the audio data using a sentiment analysis model. Specifically, it uses the DistilBERT model to obtain sentiment labels (positive, neutral, negative) for the previously saved text data.

[0671] Step 10:

[0672] The server uses an image classification model to analyze the image data. Specifically, it uses a Vision Transformer (ViT) model to obtain labels for the information the image represents (e.g., a graph of rising stock prices).

[0673] Step 11:

[0674] The server uses an emotion engine to analyze the user's text, voice, and biometric sensor data to estimate the user's emotional state and obtain emotion labels.

[0675] Step 12:

[0676] The server integrates all data analysis results, specifically, compiles emotion and image labels obtained from each data source into a single list and performs a comprehensive evaluation.

[0677] Step 13:

[0678] The server automatically generates a trading strategy based on the integrated analysis results and user sentiment data. Specifically, if the sentiment analysis results and user sentiment are stable, it generates a "BUY" strategy, and if not, it generates a "SELL" strategy.

[0679] Step 14:

[0680] The server sends the generated trading strategy to the user's terminal, notifying the user by displaying a message saying, "The current recommended trading strategy is 'BUY'."

[0681] Step 15:

[0682] The user accesses the trading platform based on the provided trading strategy and takes specific investment actions, such as "buying" or "selling" according to the displayed strategy.

[0683] In this way, the server can comprehensively analyze market data and user sentiment data, and generate and provide highly accurate trading strategies in real time, thereby significantly supporting users' investment decisions.

[0684] Example 2

[0685] 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."

[0686] It is extremely difficult for investors to make instant decisions amidst the rapid fluctuations in information and vast amounts of data in financial markets. Furthermore, because investors' emotional state has a significant impact on investment decisions, strategies that ignore emotions have limited effectiveness. Therefore, in order for investors to make faster and more accurate decisions, a system is needed that can comprehensively analyze market data and emotional data and provide investors with specific trading strategies in real time.

[0687] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0688] In this invention, the server includes means for collecting financial market data, means for performing multimodal analysis of the collected data and user emotion data, means for generating a trading strategy based on the analysis results and the user emotion data, and means for providing the generated trading strategy to investors, thereby enabling investors to grasp market trends and obtain highly accurate trading strategies that take their own emotional state into account in real time.

[0689] "Financial market data" refers to various information obtained from financial markets, including stock prices, exchange rates, interest rates, economic indicators, news articles, reports, etc.

[0690] "Multimodal analysis" is a method for integrating and analyzing data of different formats (text, audio, images, biometric sensors, etc.).

[0691] A "trading strategy" refers to decisions and plans for buying and selling in financial markets, and specifically includes recommendations for "buying" and "selling."

[0692] "Emotion data" is data that represents the psychological state of the user, and includes information obtained from text messages, voice input, biometric sensors, and the like.

[0693] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to analyze data and generate a specific outcome (e.g., a sentiment label or a trading strategy).

[0694] A "prompt sentence" is an input sentence that causes a generative AI model to perform a specific analysis or generation task.

[0695] This invention relates to a system for supporting investment analysis in financial markets, specifically an "Intelligent Market Sensor" system. This system uses generative AI models and AGI to comprehensively analyze multimodal data, including text, voice, images, and biometric sensors, to detect market trends, risk factors, and investment opportunities in real time, and provide specific trading strategies to investors.

[0696] Data collection

[0697] The server retrieves text data from a financial news API and collects audio data from a financial information service. It also retrieves image data from an image API and subjects these data to an analysis step. For example, the server may retrieve text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest. At the same time, it retrieves audio of a financial analyst's speech from https: / / api.financial-audio.com / latest and image data showing rising stock prices from https: / / api.financial-images.com / latest.

[0698] Emotional Data Collection

[0699] The server collects emotion-related data from the user's text messages, speech input, and biometric sensors. This data includes, for example, the user's text input (e.g., "What do you think about the market these days?"), speech, and biometric data such as heart rate.

[0700] Data analysis

[0701] The server analyzes the collected text data using a generative AI model (for example, OpenAI's GPT-4) to obtain an emotion label. It also analyzes voice data in the same way, converts it into text, and then obtains an emotion label. This conversion is performed using Google's Speech-to-Text API. Image data is analyzed using an image classification model such as Google's Vision API, and the content is classified as a label. For example, text data such as "The market is currently on the rise" is input into a generative AI model to create a prompt that obtains the emotion label "positive."

[0702] Emotional Data Analysis

[0703] The server uses an emotion engine to analyze data obtained from the user's text, voice, and biometric sensors to recognize the user's emotional state. For example, it identifies emotions such as "excited" from the user's heart rate and facial expression data. This recognized emotion data is used in the subsequent trading strategy generation step.

[0704] Trading Strategy Generation

[0705] The server automatically generates a trading strategy based on the analysis results and the user's emotional data. Specifically, if the market data's emotional label is positive and the user's emotional state is stable, a "buy" strategy is generated; if it is negative, a "sell" strategy is generated. The server prompts the generating AI model with the following prompt: "The current market data is positive. The user's emotional state is also stable. What trading strategy do you recommend?"

[0706] Offering to investors

[0707] The server then sends the generated trading strategy to the user's device and notifies the user. Based on the strategy, the user can access the trading platform and take specific investment actions. This process allows users to grasp market trends and risk factors in real time and make timely and accurate investment decisions.

[0708] This system comprehensively analyzes market data and user sentiment data, and generates and provides highly accurate trading strategies in real time, thereby significantly supporting investors' investment decisions.

[0709] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0710] Step 1:

[0711] Data collection

[0712] The server first obtains text data from a financial news API (e.g., https: / / api.financial-news.com / latest). It receives the response data (news article text) from the API as input and stores the text data in an internal database as output. At the same time, it collects audio data from a financial information service and obtains the audio file using an audio API (e.g., https: / / api.financial-audio.com / latest). It also obtains image data from an image API (e.g., https: / / api.financial-images.com / latest) and stores them in an internal database. Specifically, the server sends an HTTPS request and parses the JSON-formatted response to extract the necessary information.

[0713] Step 2:

[0714] Emotional Data Collection

[0715] The server collects emotion-related data from text messages entered by the user, audio from conversations, and biometric sensors. As input, it receives text messages, audio files, and biometric sensor data sent from the user's device. As output, it generates structured data formatted for analysis. For example, if a user types, "What do you think about the market these days?", the server collects the text and also captures the audio data and saves it as an audio file.

[0716] Step 3:

[0717] Data analysis

[0718] The server inputs the collected text data into a generative AI model (e.g., OpenAI's GPT-4) to obtain an emotion label. The collected text data is provided as input, and a generated emotion label (e.g., "positive" or "negative") is obtained as output. Voice data is also converted into text using voice recognition software (e.g., Google's Speech-to-Text API), which is then analyzed by the generative AI model. Image data is analyzed using an image classification model (e.g., Google's Vision API), and the content of the image is classified as a label. Specifically, the analysis results in label information such as "the market is on the rise," "financial analysts' comments are positive," and "image showing rising stock prices."

[0719] Step 4:

[0720] Emotional Data Analysis

[0721] The server uses an emotion engine to analyze data obtained from the user's text, voice, and biometric sensors to recognize the user's emotional state. The server provides the user's text data, voice data, and biometric data as input, and identifies the user's emotional state (e.g., "excited" or "calm") as output. Specifically, the server runs the text analysis engine, voice analysis engine, and biometric sensor analysis engine to evaluate the user's overall emotional state.

[0722] Step 5:

[0723] Trading Strategy Generation

[0724] The server automatically generates trading strategies based on market data and user sentiment data. It provides analysis results and sentiment data as input, and generates specific trading strategies such as "buy" or "sell" as output. This is done by inputting prompts such as "Current market data is positive. The user's sentiment state is stable. What trading strategy do you recommend?" into the generative AI model, and obtaining output from the model. For example, if the overall market data is positive, a "buy" strategy is generated.

[0725] Step 6:

[0726] Offering to investors

[0727] The server sends the generated trading strategy to the user's device and notifies the user. It provides the generated trading strategy data as input and sends a notification in the form of a message to the user's device as output. The user receives the notification and accesses a trading platform (e.g., an online securities site or a dedicated app) to perform specific investment actions. Specifically, the server sends a notification to the user's device via a REST API, and the user can make investments through the trading application.

[0728] (Application example 2)

[0729] 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."

[0730] Conventional trading strategy generation systems are limited to analyzing financial market data and do not address the optimization of sales strategies in brick-and-mortar stores. Furthermore, because strategies are not generated taking into account the emotional state of customers, it is difficult to improve customer satisfaction and sales efficiency. Therefore, there is a need for a method to optimize brick-and-mortar store sales strategies in real time, reflecting the emotional state of customers.

[0731] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0732] In this invention, the server includes means for collecting financial market data and data from a physical store, means for performing multimodal analysis of the collected data and recognizing customer emotion data, means for generating trading and sales strategies based on the analysis results and the customer emotion data, and means for providing the generated trading and sales strategies. This makes it possible to generate sales strategies based on product trends and customer emotions in the physical store in addition to analyzing financial market data, thereby improving customer satisfaction and maximizing sales.

[0733] "Financial market data" refers to information such as text, audio, images, and video related to financial markets, including stock prices, trading volume, corporate financial status, economic indicators, and news.

[0734] "In-store data" refers to information such as customer behavior, purchase history, inventory status, and promotional effectiveness in physical stores, and includes data obtained from IoT sensors and cameras.

[0735] "Multimodal analysis" is a technology that integrates and analyzes multiple data formats, such as text, audio, images, and video, and is a method of extracting useful information from data using generative AI models.

[0736] "Customer emotion data" refers to the emotional state analyzed based on data obtained from the customer's facial expressions, voice, text messages, and biometric sensors, and includes emotional categories such as positive, negative, excited, and calm.

[0737] A "trading strategy" is a plan that outlines the method or direction for executing transactions in financial markets, including buying and selling decisions.

[0738] A "sales strategy" is a plan that outlines the methods and policies for selling products in physical stores, including strengthening promotions and adjusting inventory.

[0739] A "generative AI model" is a model that uses artificial intelligence techniques to analyze collected data and is trained to perform specific tasks, such as pattern recognition and predictions in the data.

[0740] "Emotion engine" refers to technology that recognizes and labels emotions from a user's text, voice, and biometric sensors, and is used to perform data analysis based on their emotional state.

[0741] "Increased promotion" refers to increasing sales promotion activities such as advertising and discounts for specific products or services in order to promote product sales.

[0742] An "IoT sensor" is a sensor connected to the Internet that measures and collects physical environmental data and transmits the data in real time.

[0743] "Sales maximization" refers to management and strategic activities aimed at increasing the profitability of a store or business as much as possible.

[0744] The present invention is a system that collects and analyzes data from financial markets and data from physical stores, and provides users with optimal trading and sales strategies. Specific embodiments for carrying out the present invention will be described below.

[0745] System Configuration

[0746] Hardware and Software

[0747] 1. Server: A central processing unit that collects data, analyzes it, and generates strategies. It has a high-performance processor and sufficient storage.

[0748] 2. IoT Sensors: These sensors collect environmental data within physical stores, measuring temperature, humidity, movement, etc.

[0749] 3. Camera: Captures customer movements and facial expressions in the physical store and sends them to the server as image data.

[0750] 4. Voice Assistant: A device that conducts voice interactions with customers and transmits voice data to a server.

[0751] 5. Generative AI models: Machine learning models that analyze text, voice, and image data to generate investment and sales strategies.

[0752] 6. Emotion engine: Software technology for analyzing the user's emotional state and obtaining emotion labels.

[0753] Data collection methods

[0754] The server collects financial market data and data from physical stores. Financial market data includes text data, audio data, image data, and video data obtained from news APIs and financial information providers. Physical store data is collected from IoT sensors, cameras, and voice assistants.

[0755] Data Analysis Methods

[0756] The server uses generative AI models to perform multimodal analysis of the collected data: text data is analyzed using natural language processing (NLP) techniques, voice data is converted to text and then labeled with an emotion label, and image data is analyzed using image classification models to recognize the customer's emotional state.

[0757] Emotion data analysis method

[0758] The server uses an emotion engine to analyze data from users' text, voice, and biometric sensors to recognize their emotional state. For example, if a customer is expressing positive emotions, it can provide strategies based on that.

[0759] Strategy Generation Method

[0760] Based on the analysis results and customer sentiment data, the server generates trading and sales strategies, for example, if the financial market data is positive and the customer sentiment state is positive, it will suggest a "buy" trading strategy or strengthen the promotion of a specific product.

[0761] Strategy delivery methods

[0762] The server provides the generated trading and sales strategies to the users who will actually use them, and the users then take specific investment and sales actions based on these strategies.

[0763] Specific examples

[0764] For example, the server retrieves market text data from the URL "https: / / api.financial-news.com / latest" and assigns an emotion label of "positive." It also analyzes customer facial expressions through cameras in physical stores to determine whether the customer is enjoying themselves. Based on this data, the server generates a message saying, "The current recommended trading strategy is 'BUY'," and sends it to the user's device. It also suggests to the store operator to strengthen promotions of specific products.

[0765] Example prompt sentence:

[0766] "Analyze recent customer sentiment and correlate it with point-of-sale data to optimize your sales strategy."

[0767] "How can we update the product's availability and increase promotions when we recognize high customer satisfaction for a particular product?"

[0768] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0769] Step 1:

[0770] Data collection

[0771] The server collects financial market data from news APIs, financial information services, image APIs, etc. For example, it obtains text data from "https: / / api.financial-news.com / latest," audio data from "https: / / api.financial-audio.com / latest," and image data from "https: / / api.financial-images.com / latest." It also simultaneously collects data from IoT sensors, cameras, and voice assistants in physical stores. Various APIs are used as input, and the collected text, audio, and image data are obtained as output.

[0772] Step 2:

[0773] Data Preprocessing

[0774] The server converts the collected raw data into a format suitable for analysis: it standardizes text data, converts voice data into text using voice recognition technology, and preprocesses image data to standardize pixel information. It receives the collected text, voice, and image data as input, and outputs a standardized dataset.

[0775] Step 3:

[0776] Emotional Data Collection

[0777] The server collects user input text, voice, and biometric sensor data. Specifically, voice data is obtained through a voice assistant, and biometric sensor data is collected from heart rate and facial expression analysis. The input is text, voice, and biometric sensor data from the user, and the output is data that can be labeled with emotion.

[0778] Step 4:

[0779] Data analysis

[0780] The server uses a generative AI model to analyze collected text, audio, and image data to obtain emotion labels. For example, it performs natural language processing (NLP) on text data, emotion recognition on audio data, and content classification on image data. It receives a standardized dataset as input and obtains emotion labels as output.

[0781] Step 5:

[0782] Emotional Data Analysis

[0783] The server uses an emotion engine to comprehensively analyze the user's emotional data, recognize the user's emotional state from text, voice, and biometric sensor data, and generate an emotion label based on this. It receives the data collected from the user as input and obtains the user's emotional state data as output.

[0784] Step 6:

[0785] Strategy Generation

[0786] The server generates trading and sales strategies based on the analysis results and user emotion data, for example, making buying and selling decisions or proposing promotion enhancements. It receives emotion labels and analysis results as input and obtains specific strategies as output.

[0787] Step 7:

[0788] Strategy Providing

[0789] The server provides the generated trading and selling strategies to the user. Specifically, it sends a message to the user's terminal saying "The current recommended trading strategy is 'BUY'." It receives the generated strategies as input and displays a notification message on the user's terminal as output.

[0790] 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.

[0791] 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.

[0792] 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.

[0793] [Third embodiment]

[0794] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0795] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0796] 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).

[0797] 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.

[0798] 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.

[0799] 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).

[0800] 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.

[0801] 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.

[0802] 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.

[0803] 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.

[0804] 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.

[0805] 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."

[0806] This invention relates to an "Intelligent Market Sensor" system that supports investment analysis in financial markets. This system utilizes generative AI and AGI to comprehensively analyze multimodal data such as text, audio, images, and video to detect market trends, risk factors, and investment opportunities in real time, and provides investors with specific trading strategies.

[0807] Explanation of program processing

[0808] Processing data collection

[0809] The server collects the latest data from various data sources, such as financial news, audio, and images. Specifically, it retrieves the latest news articles from the financial news API, downloads audio data, and retrieves and loads image data. This collected data is then used for subsequent analysis steps.

[0810] Data analysis process

[0811] The server analyzes the collected text, audio, and image data. For text data, it performs sentiment analysis using a generative AI model (e.g., DistilBERT). For audio data, it uses a generative AI model (e.g., Wav2Vec2) to convert audio to text, and then further analyzes the text for sentiment. For image data, it analyzes it using an image classification model (e.g., Vision Transformer) to obtain the information indicated by the image as a label.

[0812] Processing of Trading Strategy Generation

[0813] The server automatically generates a trading strategy based on the analysis results. Specifically, it integrates the results of text, voice, and image analysis, and generates a strategy to "buy" if the analysis results are positive, and "sell" if they are negative. This generated strategy becomes important information to support investors' decision-making.

[0814] Processing of investor offerings

[0815] The server then sends the generated trading strategy to the user's device. The user then accesses the trading platform and executes specific investment actions based on the strategy provided. In this process, the user is able to grasp market trends and risk factors in real time, enabling them to make timely and accurate investment decisions.

[0816] Specific examples

[0817] Consider the following example as a specific scenario.

[0818] 1. Data collection

[0819] For example, the server retrieves text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest, while simultaneously retrieving audio of a financial analyst from https: / / api.financial-audio.com / latest and image data showing rising stock prices from https: / / api.financial-images.com / latest.

[0820] 2. Data Analysis

[0821] For this collected text data, the server uses a generative AI model to obtain the emotion label "positive." After converting the voice data into text using the generative AI model, the server performs a similar emotion analysis and obtains the "positive" label. Furthermore, the generative AI model is used to analyze image data, obtaining the label "stock price rise."

[0822] 3. Generating a Trading Strategy

[0823] The server aggregates all the analysis results and generates a "buy" strategy since the overall result is positive. This strategy provides useful guidance to investors.

[0824] 4. Offering to investors

[0825] The server sends a message to the user's terminal saying, "The current recommended trading strategy is 'BUY'." Based on this information, the user takes specific investment action through the trading platform.

[0826] This "Intelligent Market Sensor" system comprehensively analyzes information from a variety of data sources and generates and provides highly accurate trading strategies in real time, thereby significantly supporting investors' investment decisions.

[0827] The processing flow will be explained below.

[0828] Step 1:

[0829] The server retrieves the latest text data from the financial news API, specifically, collects JSON data of news articles from https: / / api.financial-news.com / latest.

[0830] Step 2:

[0831] The server collects audio data from the financial information service by downloading the audio file from https: / / api.financial-audio.com / latest and saving the data.

[0832] Step 3:

[0833] The server retrieves the latest financial image data from the image API by accessing the endpoint https: / / api.financial-images.com / latest and loading the retrieved image data as an image object using Image.open .

[0834] Step 4:

[0835] The server analyzes the collected text data using a sentiment analysis model. Specifically, it runs a text sentiment analysis pipeline using the DistilBERT model and outputs sentiment labels (positive, neutral, or negative) for news articles.

[0836] Step 5:

[0837] The server uses a model to convert audio data into text. Specifically, it transcribes the audio data using the Wav2Vec2 model and stores the resulting text data.

[0838] Step 6:

[0839] The server analyzes the text extracted from the audio data using a sentiment analysis model. Specifically, it uses the DistilBERT model to obtain sentiment labels (positive, neutral, negative) for the previously saved text data.

[0840] Step 7:

[0841] The server uses an image classification model to analyze the image data. Specifically, it uses a Vision Transformer (ViT) model to obtain labels for the information the image represents (e.g., a graph of rising stock prices).

[0842] Step 8:

[0843] The server integrates all data analysis results, specifically, compiles emotion and image labels obtained from each data source into a single list and performs a comprehensive evaluation.

[0844] Step 9:

[0845] The server automatically generates a trading strategy based on the integrated analysis results. Specifically, if the proportion of positive sentiment labels is high, it generates a "BUY" strategy, and if not, it generates a "SELL" strategy.

[0846] Step 10:

[0847] The server sends the generated trading strategy to the user's terminal. The user can take specific investment actions based on the provided strategy. For example, a message such as "The current recommended trading strategy is 'BUY'" is displayed on the terminal.

[0848] Example 1

[0849] 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."

[0850] In financial markets, it is extremely important for investors to grasp market trends, risk factors, and investment opportunities in real time and make accurate and prompt decisions. However, conventional systems have fragmented information collection, analysis, and strategy generation, making it difficult to comprehensively analyze information from diverse data sources. In particular, there is a need for unified processing of data in different formats, such as text, audio, and images, and for generating and providing reliable trading strategies based on the analysis results.

[0851] 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.

[0852] In this invention, the server includes a means for collecting financial market information, a means for multimodal analysis of the collected information, a means for analyzing text, voice, and image data using a generative AI model, a means for generating a trading strategy based on the analysis results, and a means for providing the generated trading strategy to investors, thereby enabling unified processing of data in different formats and providing reliable strategies in real time.

[0853] "Financial market information" refers to all data necessary to understand market trends, such as stock prices, exchange rates, economic indicators, and corporate financial information.

[0854] "Collect" refers to the act of obtaining data from designated data sources and storing it in the system.

[0855] "Multimodal analysis" refers to the simultaneous analysis of multiple forms of data, such as text, audio, and images, to obtain comprehensive information.

[0856] A "generative AI model" refers to an algorithm or system that uses artificial intelligence techniques to analyze and process data.

[0857] A "trading strategy" refers to an investor's plan or policy for buying and selling in financial markets.

[0858] "Providing to investors" means sending the generated information and strategies to the user's device or platform so that they can be used by investors.

[0859] "Sentiment analysis" refers to a technique for identifying emotions and opinions from text data and quantitatively evaluating them.

[0860] "Buy or sell recommendation" refers to a suggestion based on the results of an analysis as to whether to buy or sell a particular financial asset.

[0861] This invention relates to an "Intelligent Market Sensor" system for supporting investment analysis in financial markets. This system utilizes generative AI and AGI to comprehensively analyze multimodal data such as text, audio, images, and video to detect market trends, risk factors, and investment opportunities in real time, and provides specific trading strategies to investors.

[0862] The server collects the latest data from various data sources, such as financial news, audio, and images. For example, the server retrieves the latest news articles from a financial news API, downloads audio data, and retrieves and loads image data. This collected data is then used for subsequent analysis steps. Generative AI models (e.g., DistilBERT, Wav2Vec2, Vision Transformer) are used to analyze the collected text, audio, and image data.

[0863] First, the server performs sentiment analysis on the collected text data using a generative AI model (e.g., DistilBERT). For audio data, it converts it into text using a generative AI model (e.g., Wav2Vec2), and then further analyzes the text for sentiment analysis. For image data, it analyzes it using an image classification model (e.g., Vision Transformer), and obtains the information indicated by the image as a label.

[0864] The server then automatically generates a trading strategy based on the analysis results. Specifically, it integrates the results of text, voice, and image analysis, and generates a "buy" strategy if the analysis results are positive, and a "sell" strategy if the analysis results are negative. This generated strategy becomes important information to support investors' decision-making.

[0865] Finally, the server sends the generated trading strategy to the user's device. The user then accesses the trading platform based on the strategy and executes specific investment actions. In this process, the user is able to grasp market trends and risk factors in real time, enabling them to make timely and accurate investment decisions.

[0866] Consider the following example as a specific scenario.

[0867] 1. Data collection

[0868] For example, the server retrieves text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest. At the same time, it retrieves audio of a financial analyst from https: / / api.financial-audio.com / latest and image data showing rising stock prices from https: / / api.financial-images.com / latest.

[0869] 2. Data Analysis

[0870] The server uses a generative AI model to obtain the emotion label "positive" from the collected text data. It also converts voice data into text using the generative AI model, then performs emotion analysis and obtains the "positive" label. It also analyzes image data using the generative AI model and obtains the label "stock price rise."

[0871] 3. Generating a Trading Strategy

[0872] The server aggregates all the analysis results and generates a "buy" strategy since the overall result is positive. This strategy provides useful guidance to investors.

[0873] 4. Offering to investors

[0874] The server sends a message to the user's terminal saying, "The current recommended trading strategy is 'BUY'." Based on this information, the user takes specific investment action through the trading platform.

[0875] Here are some example prompts you can enter into your generative AI model:

[0876] 1. Prompts for text sentiment analysis

[0877] "Analyze the sentiment of the following financial news article: The market is currently trending upward."

[0878] 2. Speech-to-text prompts

[0879] "Transcribe this financial analyst's speech to text: Audio data stream"

[0880] 3. Prompt for image analysis

[0881] "Please classify the content of this image and label it appropriately: Image Data Stream"

[0882] In this way, the present invention comprehensively analyzes information from various data sources such as text, audio, and image data, and generates and provides highly accurate trading strategies in real time, thereby greatly supporting investors' investment decisions.

[0883] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0884] Step 1:

[0885] The server retrieves the latest news articles from the financial news API.

[0886] Specifically, the server sends a GET request to the URL https: / / api.financial-news.com / latest and receives news data in text format. The retrieved news articles are stored in JSON format.

[0887] Input: Financial News API URL

[0888] Data processing: GET request, saving received data in JSON format

[0889] Output: News data in JSON format

[0890] Step 2:

[0891] The server downloads the audio data.

[0892] Specifically, the server downloads audio data from https: / / api.financial-audio.com / latest, saves it in WAV format, and later converts it to text.

[0893] Input: URL of audio data

[0894] Data processing: HTTP requests, saving received data in WAV format

[0895] Output: WAV format audio data

[0896] Step 3:

[0897] The server acquires and reads the image data.

[0898] Specifically, the server downloads image data from https: / / api.financial-images.com / latest, saves it in JPEG format, and later feeds it into the image analysis model.

[0899] Input: Image data URL

[0900] Data processing: HTTP requests, saving received data in JPEG format

[0901] Output: JPEG format image data

[0902] Step 4:

[0903] The server performs sentiment analysis on the collected text data using a generative AI model.

[0904] Specifically, the server inputs text data into a generative AI model (e.g., DistilBERT) to obtain sentiment labels. This step uses the prompt sentence, "Analyze the sentiment of the following financial news article: The market is currently trending upward."

[0905] Input: JSON format text data, prompt statement

[0906] Data Processing: Sentiment Analysis with Generative AI Models

[0907] Output: Sentiment label (e.g. "positive")

[0908] Step 5:

[0909] The server converts the audio data into text using a generative AI model.

[0910] Specifically, the server inputs WAV-formatted audio data into a generative AI model (e.g., Wav2Vec2) and obtains text data. This procedure uses the prompt "Convert this financial analyst's speech into text: Audio data stream."

[0911] Input: WAV format audio data, prompt text

[0912] Data Processing: Speech-to-Text Conversion with Generative AI Models

[0913] Output: Text data

[0914] Step 6:

[0915] The server performs sentiment analysis on the text data converted from the voice data.

[0916] Specifically, the server inputs the converted text data into a generative AI model (e.g., DistilBERT) to obtain a sentiment label. This step uses the prompt sentence, "Analyze the sentiment of the following text: The market is trending upward."

[0917] Input: Text data, prompt

[0918] Data Processing: Sentiment Analysis with Generative AI Models

[0919] Output: Sentiment label (e.g. "positive")

[0920] Step 7:

[0921] The server analyzes the image data with an image classification model.

[0922] Specifically, the server inputs image data into a generative AI model (e.g., Vision Transformer) to obtain image labels. This procedure uses the prompt "Please classify the content of this image and provide an appropriate label: image data stream."

[0923] Input: JPEG image data, prompt text

[0924] Data Processing: Image Classification with Generative AI Models

[0925] Output: Image label (e.g. "Stock price rise")

[0926] Step 8:

[0927] The server integrates the results of text, voice and image analysis.

[0928] Specifically, the server integrates the results of each analysis and determines an overall emotion label. For example, if all emotion labels are "positive," the overall result is also determined to be positive.

[0929] Input: Text analysis results, voice analysis results, image analysis results

[0930] Data processing: Integrating each result and determining the overall sentiment

[0931] Output: Overall sentiment label (e.g. "positive")

[0932] Step 9:

[0933] The server generates a trading strategy based on the analysis results.

[0934] Specifically, the server automatically generates a strategy based on the integrated analysis results. For example, if the analysis results are positive, it generates a strategy to "buy," and if they are negative, it generates a strategy to "sell."

[0935] Input: Overall emotion label

[0936] Data processing: Automatic generation of strategies

[0937] Output: Trading strategy (e.g. "Buy")

[0938] Step 10:

[0939] The server transmits the generated trading strategy to the user's terminal.

[0940] As a specific operation, the server transmits the generated strategy to the user's terminal in the form of a message.

[0941] Input: Trading Strategy

[0942] Data processing: Message generation and transmission

[0943] Output: A message to the user terminal (e.g., "The current recommended trading strategy is 'BUY'")

[0944] Step 11:

[0945] Based on the provided strategy, the user accesses the trading platform and executes specific investment actions.

[0946] Specifically, the user operates the trading platform based on the message received from the server and executes the instructed investment action. For example, if a "BUY" recommendation is received, the user will purchase the corresponding stock.

[0947] Input: Trading strategy message from the server

[0948] Data processing: input to trading platform

[0949] Output: Specific investment action (e.g., purchase of stock)

[0950] (Application example 1)

[0951] 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."

[0952] Conventional financial market analysis systems rely on limited information sources for data collection and analysis, and lack data diversification to improve the accuracy of trading strategies provided to investors. On the other hand, systems that use user activity data to recommend individually optimized content also face the challenge of making appropriate recommendations based on real-time data analysis. Therefore, the present invention aims to provide highly accurate trading strategies and individually optimized content by collecting and analyzing financial market data and user activity data in a multimodal manner.

[0953] 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.

[0954] In this invention, the server includes means for collecting financial market data, means for multimodal analysis of the collected data, means for generating a trading strategy based on the analysis results, means for providing the generated trading strategy to investors, means for collecting user activity data, means for analyzing the collected user activity data to generate optimal content, and means for providing the generated content to users. This makes it possible to comprehensively analyze data related to financial markets and user activity and provide appropriate investment strategies and content in real time.

[0955] "Financial market data" refers to information related to market trends, such as stock prices, exchange rates, interest rates, and financial reports.

[0956] "Multimodal analysis" refers to a method of comprehensive analysis using multiple data formats such as text, audio, images, and video.

[0957] A "trading strategy" refers to a specific plan for determining the timing and method of buying and selling in financial markets.

[0958] "Means for providing to investors" refers to methods and techniques for communicating the generated trading strategies to investors.

[0959] "User activity data" refers to information related to a user's behavior, such as a user's browsing history, ratings, and feedback.

[0960] "Means for generating optimal content" refers to methods and technologies for recommending and creating content that is most suitable for users based on collected user activity data.

[0961] "Means for providing generated content to a user" refers to methods and technologies for providing content suitable for a user to a user's device or application.

[0962] This invention relates to a system that collects and analyzes financial market data and user activity data in a multimodal manner, and provides highly accurate trading strategies and individually optimized content.

[0963] System Configuration

[0964] 1. Hardware

[0965] Server: Responsible for data collection, analysis, trading strategy and content generation. Equipped with a high-performance processor and large memory capacity.

[0966] User devices: Smartphones, smart glasses, head-mounted displays, etc. Providing trading strategies and content to users.

[0967] 2. Software

[0968] Prediction models: DistilBERT is used for text analysis, Wav2Vec2 for speech analysis, and Vision Transformer for image analysis.

[0969] API: RESTful API for collecting financial news, audio data, image data, user activity data, etc.

[0970] Database: A database for storing collected data and analysis results.

[0971] Program processing

[0972] Data collection

[0973] The server collects financial market data and user activity data, for example, using the following APIs:

[0974] Get the latest market information from our financial news API.

[0975] Collect trending data from social media APIs.

[0976] Obtain user browsing history from the content platform API.

[0977] Data analysis

[0978] The server analyzes the collected data. Specific examples include:

[0979] The text data is subjected to sentiment analysis using the DistilBERT model to determine whether it is positive or negative.

[0980] The audio data is converted into text using the Wav2Vec2 model, and then sentiment analysis is performed.

[0981] Image data is analyzed using the Vision Transformer model, and the information indicated by the image is obtained as labels.

[0982] Trading Strategies and Content Generation

[0983] The server generates trading strategies and content based on the analysis results.

[0984] If the analysis result is positive, a trading strategy of "buy" is generated, and if it is negative, a trading strategy of "sell" is generated.

[0985] Recommend the most suitable content based on user activity data.

[0986] Provision to users

[0987] The generated trading strategies and content are sent to the user's terminal, where the user can make investment decisions and view the content based on the information provided.

[0988] It sends notifications to smartphones and smart glasses apps, displaying current trading strategy recommendations and content to users.

[0989] Specific examples

[0990] For example, the server retrieves text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest, and simultaneously collects user browsing history data from https: / / api.content-platform.com / user_activity.

[0991] The server then uses a generative AI model to generate a "positive" sentiment label for the collected text data. It then performs a similar analysis on the user's browsing history data to identify patterns of content that the user prefers.

[0992] Finally, the server integrates all the analysis results and generates a "buy" strategy since the overall results are positive. It also recommends content that the user will like and sends a notification to the user's smartphone, such as "The current recommended content is 'Latest Movie Trends'."

[0993] Prompt Sentence Examples

[0994] "Analyze users' browsing history and feedback over the past six months to recommend content that is likely to generate a positive response."

[0995] As described above, the present invention is a system that can comprehensively analyze financial market and user activity data and provide two important functions, investment strategies and content recommendations, in real time.

[0996] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0997] Step 1:

[0998] The server collects financial market data and user activity data. Specifically, it obtains the latest market information from a financial news API, collects trend-related data from a social media API, and obtains user browsing history from a content platform API. In this case, the input is financial market information and user activity data, and the output is the collected data.

[0999] Step 2:

[1000] The server analyzes the collected data. Specifically, it performs sentiment analysis on the collected text data using the DistilBERT model, converts the audio data to text using the Wav2Vec2 model, then performs sentiment analysis, and analyzes the image data using the Vision Transformer model to obtain the information indicated by the image as a label. The inputs are text data, audio data, and image data, and the output is the analysis results: emotion labels and image labels.

[1001] Step 3:

[1002] The server generates trading strategies and content based on the analysis results. For example, if the analysis results of text, audio, and images are positive, a trading strategy of "buy" is generated, and if they are negative, a trading strategy of "sell" is generated. The server also recommends optimal content based on user activity data. In this case, the input is the analysis results and user activity data, and the output is the generated trading strategy and content.

[1003] Step 4:

[1004] The server provides the generated trading strategies and content to the user's device. The server notifies the user's smartphone or smart glasses app of the recommended trading strategies and content, allowing the user to access them at any time. In this case, the input is the generated trading strategies and content, and the output is the information provided to the user.

[1005] Step 5:

[1006] Based on the provided trading strategies and content, users take appropriate investment actions and view content. Specifically, they execute trades on the trading platform based on notifications or view recommended content. In this case, the input is the generated information and the output is the user's actions.

[1007] 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.

[1008] This invention relates to an "intelligent market sensor" system that supports investment analysis in financial markets. This system utilizes generative AI and AGI to comprehensively analyze multimodal data, including text, voice, images, and video, to detect market trends, risk factors, and investment opportunities in real time and provide specific trading strategies to investors. Furthermore, this system incorporates an emotion engine that recognizes user emotions, enabling it to optimize trading strategies taking user emotions into account.

[1009] Explanation of program processing

[1010] Processing data collection

[1011] The server acquires text data from the financial news API, collects audio data from the financial information service, and acquires image data from the image API, and subjects these data to an analysis step.

[1012] Processing emotion data collection

[1013] The server collects emotion-related data from the user's text messages, voice inputs, and biometric sensors, including the user's heart rate and facial expression analysis data.

[1014] Data analysis process

[1015] The server analyzes the collected text data using a generative AI model to obtain emotion labels. It also analyzes voice data, converts it into text, and obtains emotion labels. It analyzes image data using an image classification model to classify the image content as a label.

[1016] Processing of sentiment data analysis

[1017] The server uses an emotion engine to analyze data obtained from the user's text, voice, and biometric sensors to recognize the user's emotional state, which is then used in the subsequent trading strategy generation step.

[1018] Processing of Trading Strategy Generation

[1019] The server automatically generates a trading strategy based on the analysis results and the user's emotion data. Specifically, if the emotion label of the market data is positive and the user's emotion is stable, the server generates a "buy" strategy, and if it is negative, the server generates a "sell" strategy.

[1020] Processing of investor offerings

[1021] The server then sends the generated trading strategy to the user's device and notifies the user. The user then accesses the trading platform based on the strategy provided and executes specific investment actions. In this process, the user is able to grasp market trends and risk factors in real time, enabling them to make timely and accurate investment decisions.

[1022] Specific examples

[1023] Consider the following example as a specific scenario.

[1024] 1. Data collection

[1025] For example, the server retrieves text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest, while simultaneously retrieving audio of a financial analyst from https: / / api.financial-audio.com / latest and image data showing rising stock prices from https: / / api.financial-images.com / latest.

[1026] 2. Collecting Emotional Data

[1027] The server collects the user's input text, speech, and data from biometric sensors (e.g., heart rate data). For example, if a user types, "What do you think about the market these days?", the server collects that text and speech.

[1028] 3. Data Analysis

[1029] For this collected text data, the server uses a generative AI model to obtain the emotion label "positive." After converting the voice data into text using the generative AI model, the server performs a similar emotion analysis and obtains the "positive" label. Furthermore, the generative AI model is used to analyze image data, obtaining the label "stock price rise."

[1030] 4. Emotion Data Analysis

[1031] The server uses an emotion engine to analyze the user's text, voice, and biometric sensor data, recognizing, for example, if the user indicates an emotion such as "excited."

[1032] 5. Generating a Trading Strategy

[1033] The server integrates all the analysis results and the user's emotional state, and generates a "buy" strategy since the overall results are positive and the user's emotions are stable.

[1034] 6. Offering to investors

[1035] The server sends a message to the user's terminal saying, "The current recommended trading strategy is 'BUY'." Based on this information, the user takes specific investment action through the trading platform.

[1036] This system comprehensively analyzes market data and user sentiment data, and generates and provides highly accurate trading strategies in real time, thereby significantly supporting investors' investment decisions.

[1037] The processing flow will be explained below.

[1038] Step 1:

[1039] The server retrieves the latest text data from the financial news API, specifically, collects JSON data of news articles from https: / / api.financial-news.com / latest.

[1040] Step 2:

[1041] The server collects audio data from the financial information service by downloading the audio file from https: / / api.financial-audio.com / latest and saving it.

[1042] Step 3:

[1043] The server retrieves financial image data from the image API by accessing the endpoint https: / / api.financial-images.com / latest and loading the retrieved image data as an image object using Image.open .

[1044] Step 4:

[1045] The server collects the user's input text, specifically, collects messages entered by the user in a chat window of the trading platform as a log.

[1046] Step 5:

[1047] The server collects voice input data from the user, specifically, stores voice data acquired through the voice assistant function.

[1048] Step 6:

[1049] The server collects the user's biometric sensor data, specifically recording heart rate, skin temperature, and other physiological data obtained from biometric sensors such as a smartwatch.

[1050] Step 7:

[1051] The server analyzes the collected text data using a sentiment analysis model. Specifically, it runs a text sentiment analysis pipeline using the DistilBERT model and outputs sentiment labels (positive, neutral, or negative) for news articles.

[1052] Step 8:

[1053] The server uses a model to convert audio data into text. Specifically, it transcribes the audio data using the Wav2Vec2 model and stores the resulting text data.

[1054] Step 9:

[1055] The server analyzes the text extracted from the audio data using a sentiment analysis model. Specifically, it uses the DistilBERT model to obtain sentiment labels (positive, neutral, negative) for the previously saved text data.

[1056] Step 10:

[1057] The server uses an image classification model to analyze the image data. Specifically, it uses a Vision Transformer (ViT) model to obtain labels for the information the image represents (e.g., a graph of rising stock prices).

[1058] Step 11:

[1059] The server uses an emotion engine to analyze the user's text, voice, and biometric sensor data to estimate the user's emotional state and obtain emotion labels.

[1060] Step 12:

[1061] The server integrates all data analysis results, specifically, compiles emotion and image labels obtained from each data source into a single list and performs a comprehensive evaluation.

[1062] Step 13:

[1063] The server automatically generates a trading strategy based on the integrated analysis results and user sentiment data. Specifically, if the sentiment analysis results and user sentiment are stable, it generates a "BUY" strategy, and if not, it generates a "SELL" strategy.

[1064] Step 14:

[1065] The server sends the generated trading strategy to the user's terminal, notifying the user by displaying a message saying, "The current recommended trading strategy is 'BUY'."

[1066] Step 15:

[1067] The user accesses the trading platform based on the provided trading strategy and takes specific investment actions, such as "buying" or "selling" according to the displayed strategy.

[1068] In this way, the server can comprehensively analyze market data and user sentiment data, and generate and provide highly accurate trading strategies in real time, thereby significantly supporting users' investment decisions.

[1069] Example 2

[1070] 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."

[1071] It is extremely difficult for investors to make instant decisions amidst the rapid fluctuations in information and vast amounts of data in financial markets. Furthermore, because investors' emotional state has a significant impact on investment decisions, strategies that ignore emotions have limited effectiveness. Therefore, in order for investors to make faster and more accurate decisions, a system is needed that can comprehensively analyze market data and emotional data and provide investors with specific trading strategies in real time.

[1072] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1073] In this invention, the server includes means for collecting financial market data, means for performing multimodal analysis of the collected data and user emotion data, means for generating a trading strategy based on the analysis results and the user emotion data, and means for providing the generated trading strategy to investors, thereby enabling investors to grasp market trends and obtain highly accurate trading strategies that take their own emotional state into account in real time.

[1074] "Financial market data" refers to various information obtained from financial markets, including stock prices, exchange rates, interest rates, economic indicators, news articles, reports, etc.

[1075] "Multimodal analysis" is a method for integrating and analyzing data of different formats (text, audio, images, biometric sensors, etc.).

[1076] A "trading strategy" refers to decisions and plans for buying and selling in financial markets, and specifically includes recommendations for "buying" and "selling."

[1077] "Emotion data" is data that represents the psychological state of the user, and includes information obtained from text messages, voice input, biometric sensors, and the like.

[1078] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to analyze data and generate a specific outcome (e.g., a sentiment label or a trading strategy).

[1079] A "prompt sentence" is an input sentence that causes a generative AI model to perform a specific analysis or generation task.

[1080] This invention relates to a system for supporting investment analysis in financial markets, specifically an "Intelligent Market Sensor" system. This system uses generative AI models and AGI to comprehensively analyze multimodal data, including text, voice, images, and biometric sensors, to detect market trends, risk factors, and investment opportunities in real time, and provide specific trading strategies to investors.

[1081] Data collection

[1082] The server retrieves text data from a financial news API and collects audio data from a financial information service. It also retrieves image data from an image API and subjects these data to an analysis step. For example, the server may retrieve text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest. At the same time, it retrieves audio of a financial analyst's speech from https: / / api.financial-audio.com / latest and image data showing rising stock prices from https: / / api.financial-images.com / latest.

[1083] Emotional Data Collection

[1084] The server collects emotion-related data from the user's text messages, speech input, and biometric sensors. This data includes, for example, the user's text input (e.g., "What do you think about the market these days?"), speech, and biometric data such as heart rate.

[1085] Data analysis

[1086] The server analyzes the collected text data using a generative AI model (for example, OpenAI's GPT-4) to obtain an emotion label. It also analyzes voice data in the same way, converts it into text, and then obtains an emotion label. This conversion is performed using Google's Speech-to-Text API. Image data is analyzed using an image classification model such as Google's Vision API, and the content is classified as a label. For example, text data such as "The market is currently on the rise" is input into a generative AI model to create a prompt that obtains the emotion label "positive."

[1087] Emotional Data Analysis

[1088] The server uses an emotion engine to analyze data obtained from the user's text, voice, and biometric sensors to recognize the user's emotional state. For example, it identifies emotions such as "excited" from the user's heart rate and facial expression data. This recognized emotion data is used in the subsequent trading strategy generation step.

[1089] Trading Strategy Generation

[1090] The server automatically generates a trading strategy based on the analysis results and the user's emotional data. Specifically, if the market data's emotional label is positive and the user's emotional state is stable, a "buy" strategy is generated; if it is negative, a "sell" strategy is generated. The server prompts the generating AI model with the following prompt: "The current market data is positive. The user's emotional state is also stable. What trading strategy do you recommend?"

[1091] Offering to investors

[1092] The server then sends the generated trading strategy to the user's device and notifies the user. Based on the strategy, the user can access the trading platform and take specific investment actions. This process allows users to grasp market trends and risk factors in real time and make timely and accurate investment decisions.

[1093] This system comprehensively analyzes market data and user sentiment data, and generates and provides highly accurate trading strategies in real time, thereby significantly supporting investors' investment decisions.

[1094] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1095] Step 1:

[1096] Data collection

[1097] The server first obtains text data from a financial news API (e.g., https: / / api.financial-news.com / latest). It receives the response data (news article text) from the API as input and stores the text data in an internal database as output. At the same time, it collects audio data from a financial information service and obtains the audio file using an audio API (e.g., https: / / api.financial-audio.com / latest). It also obtains image data from an image API (e.g., https: / / api.financial-images.com / latest) and stores them in an internal database. Specifically, the server sends an HTTPS request and parses the JSON-formatted response to extract the necessary information.

[1098] Step 2:

[1099] Emotional Data Collection

[1100] The server collects emotion-related data from text messages entered by the user, audio from conversations, and biometric sensors. As input, it receives text messages, audio files, and biometric sensor data sent from the user's device. As output, it generates structured data formatted for analysis. For example, if a user types, "What do you think about the market these days?", the server collects the text and also captures the audio data and saves it as an audio file.

[1101] Step 3:

[1102] Data analysis

[1103] The server inputs the collected text data into a generative AI model (e.g., OpenAI's GPT-4) to obtain an emotion label. The collected text data is provided as input, and a generated emotion label (e.g., "positive" or "negative") is obtained as output. Voice data is also converted into text using voice recognition software (e.g., Google's Speech-to-Text API), which is then analyzed by the generative AI model. Image data is analyzed using an image classification model (e.g., Google's Vision API), and the content of the image is classified as a label. Specifically, the analysis results in label information such as "the market is on the rise," "financial analysts' comments are positive," and "image showing rising stock prices."

[1104] Step 4:

[1105] Emotional Data Analysis

[1106] The server uses an emotion engine to analyze data obtained from the user's text, voice, and biometric sensors to recognize the user's emotional state. The server provides the user's text data, voice data, and biometric data as input, and identifies the user's emotional state (e.g., "excited" or "calm") as output. Specifically, the server runs the text analysis engine, voice analysis engine, and biometric sensor analysis engine to evaluate the user's overall emotional state.

[1107] Step 5:

[1108] Trading Strategy Generation

[1109] The server automatically generates trading strategies based on market data and user sentiment data. It provides analysis results and sentiment data as input, and generates specific trading strategies such as "buy" or "sell" as output. This is done by inputting prompts such as "Current market data is positive. The user's sentiment state is stable. What trading strategy do you recommend?" into the generative AI model, and obtaining output from the model. For example, if the overall market data is positive, a "buy" strategy is generated.

[1110] Step 6:

[1111] Offering to investors

[1112] The server sends the generated trading strategy to the user's device and notifies the user. It provides the generated trading strategy data as input and sends a notification in the form of a message to the user's device as output. The user receives the notification and accesses a trading platform (e.g., an online securities site or a dedicated app) to perform specific investment actions. Specifically, the server sends a notification to the user's device via a REST API, and the user can make investments through the trading application.

[1113] (Application example 2)

[1114] 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."

[1115] Conventional trading strategy generation systems are limited to analyzing financial market data and do not address the optimization of sales strategies in brick-and-mortar stores. Furthermore, because strategies are not generated taking into account the emotional state of customers, it is difficult to improve customer satisfaction and sales efficiency. Therefore, there is a need for a method to optimize brick-and-mortar store sales strategies in real time, reflecting the emotional state of customers.

[1116] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1117] In this invention, the server includes means for collecting financial market data and data from a physical store, means for performing multimodal analysis of the collected data and recognizing customer emotion data, means for generating trading and sales strategies based on the analysis results and the customer emotion data, and means for providing the generated trading and sales strategies. This makes it possible to generate sales strategies based on product trends and customer emotions in the physical store in addition to analyzing financial market data, thereby improving customer satisfaction and maximizing sales.

[1118] "Financial market data" refers to information such as text, audio, images, and video related to financial markets, including stock prices, trading volume, corporate financial status, economic indicators, and news.

[1119] "In-store data" refers to information such as customer behavior, purchase history, inventory status, and promotional effectiveness in physical stores, and includes data obtained from IoT sensors and cameras.

[1120] "Multimodal analysis" is a technology that integrates and analyzes multiple data formats, such as text, audio, images, and video, and is a method of extracting useful information from data using generative AI models.

[1121] "Customer emotion data" refers to the emotional state analyzed based on data obtained from the customer's facial expressions, voice, text messages, and biometric sensors, and includes emotional categories such as positive, negative, excited, and calm.

[1122] A "trading strategy" is a plan that outlines the method or direction for executing transactions in financial markets, including buying and selling decisions.

[1123] A "sales strategy" is a plan that outlines the methods and policies for selling products in physical stores, including strengthening promotions and adjusting inventory.

[1124] A "generative AI model" is a model that uses artificial intelligence techniques to analyze collected data and is trained to perform specific tasks, such as pattern recognition and predictions in the data.

[1125] "Emotion engine" refers to technology that recognizes and labels emotions from a user's text, voice, and biometric sensors, and is used to perform data analysis based on their emotional state.

[1126] "Increased promotion" refers to increasing sales promotion activities such as advertising and discounts for specific products or services in order to promote product sales.

[1127] An "IoT sensor" is a sensor connected to the Internet that measures and collects physical environmental data and transmits the data in real time.

[1128] "Sales maximization" refers to management and strategic activities aimed at increasing the profitability of a store or business as much as possible.

[1129] The present invention is a system that collects and analyzes data from financial markets and data from physical stores, and provides users with optimal trading and sales strategies. Specific embodiments for carrying out the present invention will be described below.

[1130] System Configuration

[1131] Hardware and Software

[1132] 1. Server: A central processing unit that collects data, analyzes it, and generates strategies. It has a high-performance processor and sufficient storage.

[1133] 2. IoT Sensors: These sensors collect environmental data within physical stores, measuring temperature, humidity, movement, etc.

[1134] 3. Camera: Captures customer movements and facial expressions in the physical store and sends them to the server as image data.

[1135] 4. Voice Assistant: A device that conducts voice interactions with customers and transmits voice data to a server.

[1136] 5. Generative AI models: Machine learning models that analyze text, voice, and image data to generate investment and sales strategies.

[1137] 6. Emotion engine: Software technology for analyzing the user's emotional state and obtaining emotion labels.

[1138] Data collection methods

[1139] The server collects financial market data and data from physical stores. Financial market data includes text data, audio data, image data, and video data obtained from news APIs and financial information providers. Physical store data is collected from IoT sensors, cameras, and voice assistants.

[1140] Data Analysis Methods

[1141] The server uses generative AI models to perform multimodal analysis of the collected data: text data is analyzed using natural language processing (NLP) techniques, voice data is converted to text and then labeled with an emotion label, and image data is analyzed using image classification models to recognize the customer's emotional state.

[1142] Emotion data analysis method

[1143] The server uses an emotion engine to analyze data from users' text, voice, and biometric sensors to recognize their emotional state. For example, if a customer is expressing positive emotions, it can provide strategies based on that.

[1144] Strategy Generation Method

[1145] Based on the analysis results and customer sentiment data, the server generates trading and sales strategies, for example, if the financial market data is positive and the customer sentiment state is positive, it will suggest a "buy" trading strategy or strengthen the promotion of a specific product.

[1146] Strategy delivery methods

[1147] The server provides the generated trading and sales strategies to the users who will actually use them, and the users then take specific investment and sales actions based on these strategies.

[1148] Specific examples

[1149] For example, the server retrieves market text data from the URL "https: / / api.financial-news.com / latest" and assigns an emotion label of "positive." It also analyzes customer facial expressions through cameras in physical stores to determine whether the customer is enjoying themselves. Based on this data, the server generates a message saying, "The current recommended trading strategy is 'BUY'," and sends it to the user's device. It also suggests to the store operator to strengthen promotions of specific products.

[1150] Example prompt sentence:

[1151] "Analyze recent customer sentiment and correlate it with point-of-sale data to optimize your sales strategy."

[1152] "How can we update the product's availability and increase promotions when we recognize high customer satisfaction for a particular product?"

[1153] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1154] Step 1:

[1155] Data collection

[1156] The server collects financial market data from news APIs, financial information services, image APIs, etc. For example, it obtains text data from "https: / / api.financial-news.com / latest," audio data from "https: / / api.financial-audio.com / latest," and image data from "https: / / api.financial-images.com / latest." It also simultaneously collects data from IoT sensors, cameras, and voice assistants in physical stores. Various APIs are used as input, and the collected text, audio, and image data are obtained as output.

[1157] Step 2:

[1158] Data Preprocessing

[1159] The server converts the collected raw data into a format suitable for analysis: it standardizes text data, converts voice data into text using voice recognition technology, and preprocesses image data to standardize pixel information. It receives the collected text, voice, and image data as input, and outputs a standardized dataset.

[1160] Step 3:

[1161] Emotional Data Collection

[1162] The server collects user input text, voice, and biometric sensor data. Specifically, voice data is obtained through a voice assistant, and biometric sensor data is collected from heart rate and facial expression analysis. The input is text, voice, and biometric sensor data from the user, and the output is data that can be labeled with emotion.

[1163] Step 4:

[1164] Data analysis

[1165] The server uses a generative AI model to analyze collected text, audio, and image data to obtain emotion labels. For example, it performs natural language processing (NLP) on text data, emotion recognition on audio data, and content classification on image data. It receives a standardized dataset as input and obtains emotion labels as output.

[1166] Step 5:

[1167] Emotional Data Analysis

[1168] The server uses an emotion engine to comprehensively analyze the user's emotional data, recognize the user's emotional state from text, voice, and biometric sensor data, and generate an emotion label based on this. It receives the data collected from the user as input and obtains the user's emotional state data as output.

[1169] Step 6:

[1170] Strategy Generation

[1171] The server generates trading and sales strategies based on the analysis results and user emotion data, for example, making buying and selling decisions or proposing promotion enhancements. It receives emotion labels and analysis results as input and obtains specific strategies as output.

[1172] Step 7:

[1173] Strategy Providing

[1174] The server provides the generated trading and selling strategies to the user. Specifically, it sends a message to the user's terminal saying "The current recommended trading strategy is 'BUY'." It receives the generated strategies as input and displays a notification message on the user's terminal as output.

[1175] 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.

[1176] 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.

[1177] 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.

[1178] [Fourth embodiment]

[1179] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1180] 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.

[1181] 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).

[1182] 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.

[1183] 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.

[1184] 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).

[1185] 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.

[1186] 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.

[1187] 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.

[1188] 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.

[1189] 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.

[1190] 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.

[1191] 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."

[1192] This invention relates to an "Intelligent Market Sensor" system that supports investment analysis in financial markets. This system utilizes generative AI and AGI to comprehensively analyze multimodal data such as text, audio, images, and video to detect market trends, risk factors, and investment opportunities in real time, and provides investors with specific trading strategies.

[1193] Explanation of program processing

[1194] Processing data collection

[1195] The server collects the latest data from various data sources, such as financial news, audio, and images. Specifically, it retrieves the latest news articles from the financial news API, downloads audio data, and retrieves and loads image data. This collected data is then used for subsequent analysis steps.

[1196] Data analysis process

[1197] The server analyzes the collected text, audio, and image data. For text data, it performs sentiment analysis using a generative AI model (e.g., DistilBERT). For audio data, it uses a generative AI model (e.g., Wav2Vec2) to convert audio to text, and then further analyzes the text for sentiment. For image data, it analyzes it using an image classification model (e.g., Vision Transformer) to obtain the information indicated by the image as a label.

[1198] Processing of Trading Strategy Generation

[1199] The server automatically generates a trading strategy based on the analysis results. Specifically, it integrates the results of text, voice, and image analysis, and generates a strategy to "buy" if the analysis results are positive, and "sell" if they are negative. This generated strategy becomes important information to support investors' decision-making.

[1200] Processing of investor offerings

[1201] The server then sends the generated trading strategy to the user's device. The user then accesses the trading platform and executes specific investment actions based on the strategy provided. In this process, the user is able to grasp market trends and risk factors in real time, enabling them to make timely and accurate investment decisions.

[1202] Specific examples

[1203] Consider the following example as a specific scenario.

[1204] 1. Data collection

[1205] For example, the server retrieves text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest, while simultaneously retrieving audio of a financial analyst from https: / / api.financial-audio.com / latest and image data showing rising stock prices from https: / / api.financial-images.com / latest.

[1206] 2. Data Analysis

[1207] For this collected text data, the server uses a generative AI model to obtain the emotion label "positive." After converting the voice data into text using the generative AI model, the server performs a similar emotion analysis and obtains the "positive" label. Furthermore, the generative AI model is used to analyze image data, obtaining the label "stock price rise."

[1208] 3. Generating a Trading Strategy

[1209] The server aggregates all the analysis results and generates a "buy" strategy since the overall result is positive. This strategy provides useful guidance to investors.

[1210] 4. Offering to investors

[1211] The server sends a message to the user's terminal saying, "The current recommended trading strategy is 'BUY'." Based on this information, the user takes specific investment action through the trading platform.

[1212] This "Intelligent Market Sensor" system comprehensively analyzes information from a variety of data sources and generates and provides highly accurate trading strategies in real time, thereby significantly supporting investors' investment decisions.

[1213] The processing flow will be explained below.

[1214] Step 1:

[1215] The server retrieves the latest text data from the financial news API, specifically, collects JSON data of news articles from https: / / api.financial-news.com / latest.

[1216] Step 2:

[1217] The server collects audio data from the financial information service by downloading the audio file from https: / / api.financial-audio.com / latest and saving the data.

[1218] Step 3:

[1219] The server retrieves the latest financial image data from the image API by accessing the endpoint https: / / api.financial-images.com / latest and loading the retrieved image data as an image object using Image.open .

[1220] Step 4:

[1221] The server analyzes the collected text data using a sentiment analysis model. Specifically, it runs a text sentiment analysis pipeline using the DistilBERT model and outputs sentiment labels (positive, neutral, or negative) for news articles.

[1222] Step 5:

[1223] The server uses a model to convert audio data into text. Specifically, it transcribes the audio data using the Wav2Vec2 model and stores the resulting text data.

[1224] Step 6:

[1225] The server analyzes the text extracted from the audio data using a sentiment analysis model. Specifically, it uses the DistilBERT model to obtain sentiment labels (positive, neutral, negative) for the previously saved text data.

[1226] Step 7:

[1227] The server uses an image classification model to analyze the image data. Specifically, it uses a Vision Transformer (ViT) model to obtain labels for the information the image represents (e.g., a graph of rising stock prices).

[1228] Step 8:

[1229] The server integrates all data analysis results, specifically, compiles emotion and image labels obtained from each data source into a single list and performs a comprehensive evaluation.

[1230] Step 9:

[1231] The server automatically generates a trading strategy based on the integrated analysis results. Specifically, if the proportion of positive sentiment labels is high, it generates a "BUY" strategy, and if not, it generates a "SELL" strategy.

[1232] Step 10:

[1233] The server sends the generated trading strategy to the user's terminal. The user can take specific investment actions based on the provided strategy. For example, a message such as "The current recommended trading strategy is 'BUY'" is displayed on the terminal.

[1234] Example 1

[1235] 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."

[1236] In financial markets, it is extremely important for investors to grasp market trends, risk factors, and investment opportunities in real time and make accurate and prompt decisions. However, conventional systems have fragmented information collection, analysis, and strategy generation, making it difficult to comprehensively analyze information from diverse data sources. In particular, there is a need for unified processing of data in different formats, such as text, audio, and images, and for generating and providing reliable trading strategies based on the analysis results.

[1237] 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.

[1238] In this invention, the server includes a means for collecting financial market information, a means for multimodal analysis of the collected information, a means for analyzing text, voice, and image data using a generative AI model, a means for generating a trading strategy based on the analysis results, and a means for providing the generated trading strategy to investors, thereby enabling unified processing of data in different formats and providing reliable strategies in real time.

[1239] "Financial market information" refers to all data necessary to understand market trends, such as stock prices, exchange rates, economic indicators, and corporate financial information.

[1240] "Collect" refers to the act of obtaining data from designated data sources and storing it in the system.

[1241] "Multimodal analysis" refers to the simultaneous analysis of multiple forms of data, such as text, audio, and images, to obtain comprehensive information.

[1242] A "generative AI model" refers to an algorithm or system that uses artificial intelligence techniques to analyze and process data.

[1243] A "trading strategy" refers to an investor's plan or policy for buying and selling in financial markets.

[1244] "Providing to investors" means sending the generated information and strategies to the user's device or platform so that they can be used by investors.

[1245] "Sentiment analysis" refers to a technique for identifying emotions and opinions from text data and quantitatively evaluating them.

[1246] "Buy or sell recommendation" refers to a suggestion based on the results of an analysis as to whether to buy or sell a particular financial asset.

[1247] This invention relates to an "Intelligent Market Sensor" system for supporting investment analysis in financial markets. This system utilizes generative AI and AGI to comprehensively analyze multimodal data such as text, audio, images, and video to detect market trends, risk factors, and investment opportunities in real time, and provides specific trading strategies to investors.

[1248] The server collects the latest data from various data sources, such as financial news, audio, and images. For example, the server retrieves the latest news articles from a financial news API, downloads audio data, and retrieves and loads image data. This collected data is then used for subsequent analysis steps. Generative AI models (e.g., DistilBERT, Wav2Vec2, Vision Transformer) are used to analyze the collected text, audio, and image data.

[1249] First, the server performs sentiment analysis on the collected text data using a generative AI model (e.g., DistilBERT). For audio data, it converts it into text using a generative AI model (e.g., Wav2Vec2), and then further analyzes the text for sentiment analysis. For image data, it analyzes it using an image classification model (e.g., Vision Transformer), and obtains the information indicated by the image as a label.

[1250] The server then automatically generates a trading strategy based on the analysis results. Specifically, it integrates the results of text, voice, and image analysis, and generates a "buy" strategy if the analysis results are positive, and a "sell" strategy if the analysis results are negative. This generated strategy becomes important information to support investors' decision-making.

[1251] Finally, the server sends the generated trading strategy to the user's device. The user then accesses the trading platform based on the strategy and executes specific investment actions. In this process, the user is able to grasp market trends and risk factors in real time, enabling them to make timely and accurate investment decisions.

[1252] Consider the following example as a specific scenario.

[1253] 1. Data collection

[1254] For example, the server retrieves text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest. At the same time, it retrieves audio of a financial analyst from https: / / api.financial-audio.com / latest and image data showing rising stock prices from https: / / api.financial-images.com / latest.

[1255] 2. Data Analysis

[1256] The server uses a generative AI model to obtain the emotion label "positive" from the collected text data. It also converts voice data into text using the generative AI model, then performs emotion analysis and obtains the "positive" label. It also analyzes image data using the generative AI model and obtains the label "stock price rise."

[1257] 3. Generating a Trading Strategy

[1258] The server aggregates all the analysis results and generates a "buy" strategy since the overall result is positive. This strategy provides useful guidance to investors.

[1259] 4. Offering to investors

[1260] The server sends a message to the user's terminal saying, "The current recommended trading strategy is 'BUY'." Based on this information, the user takes specific investment action through the trading platform.

[1261] Here are some example prompts you can enter into your generative AI model:

[1262] 1. Prompts for text sentiment analysis

[1263] "Analyze the sentiment of the following financial news article: The market is currently trending upward."

[1264] 2. Speech-to-text prompts

[1265] "Transcribe this financial analyst's speech to text: Audio data stream"

[1266] 3. Prompt for image analysis

[1267] "Please classify the content of this image and label it appropriately: Image Data Stream"

[1268] In this way, the present invention comprehensively analyzes information from various data sources such as text, audio, and image data, and generates and provides highly accurate trading strategies in real time, thereby greatly supporting investors' investment decisions.

[1269] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1270] Step 1:

[1271] The server retrieves the latest news articles from the financial news API.

[1272] Specifically, the server sends a GET request to the URL https: / / api.financial-news.com / latest and receives news data in text format. The retrieved news articles are stored in JSON format.

[1273] Input: Financial News API URL

[1274] Data processing: GET request, saving received data in JSON format

[1275] Output: News data in JSON format

[1276] Step 2:

[1277] The server downloads the audio data.

[1278] Specifically, the server downloads audio data from https: / / api.financial-audio.com / latest, saves it in WAV format, and later converts it to text.

[1279] Input: URL of audio data

[1280] Data processing: HTTP requests, saving received data in WAV format

[1281] Output: WAV format audio data

[1282] Step 3:

[1283] The server acquires and reads the image data.

[1284] Specifically, the server downloads image data from https: / / api.financial-images.com / latest, saves it in JPEG format, and later feeds it into the image analysis model.

[1285] Input: Image data URL

[1286] Data processing: HTTP requests, saving received data in JPEG format

[1287] Output: JPEG format image data

[1288] Step 4:

[1289] The server performs sentiment analysis on the collected text data using a generative AI model.

[1290] Specifically, the server inputs text data into a generative AI model (e.g., DistilBERT) to obtain sentiment labels. This step uses the prompt sentence, "Analyze the sentiment of the following financial news article: The market is currently trending upward."

[1291] Input: JSON format text data, prompt statement

[1292] Data Processing: Sentiment Analysis with Generative AI Models

[1293] Output: Sentiment label (e.g. "positive")

[1294] Step 5:

[1295] The server converts the audio data into text using a generative AI model.

[1296] Specifically, the server inputs WAV-formatted audio data into a generative AI model (e.g., Wav2Vec2) and obtains text data. This procedure uses the prompt "Convert this financial analyst's speech into text: Audio data stream."

[1297] Input: WAV format audio data, prompt text

[1298] Data Processing: Speech-to-Text Conversion with Generative AI Models

[1299] Output: Text data

[1300] Step 6:

[1301] The server performs sentiment analysis on the text data converted from the voice data.

[1302] Specifically, the server inputs the converted text data into a generative AI model (e.g., DistilBERT) to obtain a sentiment label. This step uses the prompt sentence, "Analyze the sentiment of the following text: The market is trending upward."

[1303] Input: Text data, prompt

[1304] Data Processing: Sentiment Analysis with Generative AI Models

[1305] Output: Sentiment label (e.g. "positive")

[1306] Step 7:

[1307] The server analyzes the image data with an image classification model.

[1308] Specifically, the server inputs image data into a generative AI model (e.g., Vision Transformer) to obtain image labels. This procedure uses the prompt "Please classify the content of this image and provide an appropriate label: image data stream."

[1309] Input: JPEG image data, prompt text

[1310] Data Processing: Image Classification with Generative AI Models

[1311] Output: Image label (e.g. "Stock price rise")

[1312] Step 8:

[1313] The server integrates the results of text, voice and image analysis.

[1314] Specifically, the server integrates the results of each analysis and determines an overall emotion label. For example, if all emotion labels are "positive," the overall result is also determined to be positive.

[1315] Input: Text analysis results, voice analysis results, image analysis results

[1316] Data processing: Integrating each result and determining the overall sentiment

[1317] Output: Overall sentiment label (e.g. "positive")

[1318] Step 9:

[1319] The server generates a trading strategy based on the analysis results.

[1320] Specifically, the server automatically generates a strategy based on the integrated analysis results. For example, if the analysis results are positive, it generates a strategy to "buy," and if they are negative, it generates a strategy to "sell."

[1321] Input: Overall emotion label

[1322] Data processing: Automatic generation of strategies

[1323] Output: Trading strategy (e.g. "Buy")

[1324] Step 10:

[1325] The server transmits the generated trading strategy to the user's terminal.

[1326] As a specific operation, the server transmits the generated strategy to the user's terminal in the form of a message.

[1327] Input: Trading Strategy

[1328] Data processing: Message generation and transmission

[1329] Output: A message to the user terminal (e.g., "The current recommended trading strategy is 'BUY'")

[1330] Step 11:

[1331] Based on the provided strategy, the user accesses the trading platform and executes specific investment actions.

[1332] Specifically, the user operates the trading platform based on the message received from the server and executes the instructed investment action. For example, if a "BUY" recommendation is received, the user will purchase the corresponding stock.

[1333] Input: Trading strategy message from the server

[1334] Data processing: input to trading platform

[1335] Output: Specific investment action (e.g., purchase of stock)

[1336] (Application example 1)

[1337] 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."

[1338] Conventional financial market analysis systems rely on limited information sources for data collection and analysis, and lack data diversification to improve the accuracy of trading strategies provided to investors. On the other hand, systems that use user activity data to recommend individually optimized content also face the challenge of making appropriate recommendations based on real-time data analysis. Therefore, the present invention aims to provide highly accurate trading strategies and individually optimized content by collecting and analyzing financial market data and user activity data in a multimodal manner.

[1339] 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.

[1340] In this invention, the server includes means for collecting financial market data, means for multimodal analysis of the collected data, means for generating a trading strategy based on the analysis results, means for providing the generated trading strategy to investors, means for collecting user activity data, means for analyzing the collected user activity data to generate optimal content, and means for providing the generated content to users. This makes it possible to comprehensively analyze data related to financial markets and user activity and provide appropriate investment strategies and content in real time.

[1341] "Financial market data" refers to information related to market trends, such as stock prices, exchange rates, interest rates, and financial reports.

[1342] "Multimodal analysis" refers to a method of comprehensive analysis using multiple data formats such as text, audio, images, and video.

[1343] A "trading strategy" refers to a specific plan for determining the timing and method of buying and selling in financial markets.

[1344] "Means for providing to investors" refers to methods and techniques for communicating the generated trading strategies to investors.

[1345] "User activity data" refers to information related to a user's behavior, such as a user's browsing history, ratings, and feedback.

[1346] "Means for generating optimal content" refers to methods and technologies for recommending and creating content that is most suitable for users based on collected user activity data.

[1347] "Means for providing generated content to a user" refers to methods and technologies for providing content suitable for a user to a user's device or application.

[1348] This invention relates to a system that collects and analyzes financial market data and user activity data in a multimodal manner, and provides highly accurate trading strategies and individually optimized content.

[1349] System Configuration

[1350] 1. Hardware

[1351] Server: Responsible for data collection, analysis, trading strategy and content generation. Equipped with a high-performance processor and large memory capacity.

[1352] User devices: Smartphones, smart glasses, head-mounted displays, etc. Providing trading strategies and content to users.

[1353] 2. Software

[1354] Prediction models: DistilBERT is used for text analysis, Wav2Vec2 for speech analysis, and Vision Transformer for image analysis.

[1355] API: RESTful API for collecting financial news, audio data, image data, user activity data, etc.

[1356] Database: A database for storing collected data and analysis results.

[1357] Program processing

[1358] Data collection

[1359] The server collects financial market data and user activity data, for example, using the following APIs:

[1360] Get the latest market information from our financial news API.

[1361] Collect trending data from social media APIs.

[1362] Obtain user browsing history from the content platform API.

[1363] Data analysis

[1364] The server analyzes the collected data. Specific examples include:

[1365] The text data is subjected to sentiment analysis using the DistilBERT model to determine whether it is positive or negative.

[1366] The audio data is converted into text using the Wav2Vec2 model, and then sentiment analysis is performed.

[1367] Image data is analyzed using the Vision Transformer model, and the information indicated by the image is obtained as labels.

[1368] Trading Strategies and Content Generation

[1369] The server generates trading strategies and content based on the analysis results.

[1370] If the analysis result is positive, a trading strategy of "buy" is generated, and if it is negative, a trading strategy of "sell" is generated.

[1371] Recommend the most suitable content based on user activity data.

[1372] Provision to users

[1373] The generated trading strategies and content are sent to the user's terminal, where the user can make investment decisions and view the content based on the information provided.

[1374] It sends notifications to smartphones and smart glasses apps, displaying current trading strategy recommendations and content to users.

[1375] Specific examples

[1376] For example, the server retrieves text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest, and simultaneously collects user browsing history data from https: / / api.content-platform.com / user_activity.

[1377] The server then uses a generative AI model to generate a "positive" sentiment label for the collected text data. It then performs a similar analysis on the user's browsing history data to identify patterns of content that the user prefers.

[1378] Finally, the server integrates all the analysis results and generates a "buy" strategy since the overall results are positive. It also recommends content that the user will like and sends a notification to the user's smartphone, such as "The current recommended content is 'Latest Movie Trends'."

[1379] Prompt Sentence Examples

[1380] "Analyze users' browsing history and feedback over the past six months to recommend content that is likely to generate a positive response."

[1381] As described above, the present invention is a system that can comprehensively analyze financial market and user activity data and provide two important functions, investment strategies and content recommendations, in real time.

[1382] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1383] Step 1:

[1384] The server collects financial market data and user activity data. Specifically, it obtains the latest market information from a financial news API, collects trend-related data from a social media API, and obtains user browsing history from a content platform API. In this case, the input is financial market information and user activity data, and the output is the collected data.

[1385] Step 2:

[1386] The server analyzes the collected data. Specifically, it performs sentiment analysis on the collected text data using the DistilBERT model, converts the audio data to text using the Wav2Vec2 model, then performs sentiment analysis, and analyzes the image data using the Vision Transformer model to obtain the information indicated by the image as a label. The inputs are text data, audio data, and image data, and the output is the analysis results: emotion labels and image labels.

[1387] Step 3:

[1388] The server generates trading strategies and content based on the analysis results. For example, if the analysis results of text, audio, and images are positive, a trading strategy of "buy" is generated, and if they are negative, a trading strategy of "sell" is generated. The server also recommends optimal content based on user activity data. In this case, the input is the analysis results and user activity data, and the output is the generated trading strategy and content.

[1389] Step 4:

[1390] The server provides the generated trading strategies and content to the user's device. The server notifies the user's smartphone or smart glasses app of the recommended trading strategies and content, allowing the user to access them at any time. In this case, the input is the generated trading strategies and content, and the output is the information provided to the user.

[1391] Step 5:

[1392] Based on the provided trading strategies and content, users take appropriate investment actions and view content. Specifically, they execute trades on the trading platform based on notifications or view recommended content. In this case, the input is the generated information and the output is the user's actions.

[1393] 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.

[1394] This invention relates to an "intelligent market sensor" system that supports investment analysis in financial markets. This system utilizes generative AI and AGI to comprehensively analyze multimodal data, including text, voice, images, and video, to detect market trends, risk factors, and investment opportunities in real time and provide specific trading strategies to investors. Furthermore, this system incorporates an emotion engine that recognizes user emotions, enabling it to optimize trading strategies taking user emotions into account.

[1395] Explanation of program processing

[1396] Processing data collection

[1397] The server acquires text data from the financial news API, collects audio data from the financial information service, and acquires image data from the image API, and subjects these data to an analysis step.

[1398] Processing emotion data collection

[1399] The server collects emotion-related data from the user's text messages, voice inputs, and biometric sensors, including the user's heart rate and facial expression analysis data.

[1400] Data analysis process

[1401] The server analyzes the collected text data using a generative AI model to obtain emotion labels. It also analyzes voice data, converts it into text, and obtains emotion labels. It analyzes image data using an image classification model to classify the image content as a label.

[1402] Processing of sentiment data analysis

[1403] The server uses an emotion engine to analyze data obtained from the user's text, voice, and biometric sensors to recognize the user's emotional state, which is then used in the subsequent trading strategy generation step.

[1404] Processing of Trading Strategy Generation

[1405] The server automatically generates a trading strategy based on the analysis results and the user's emotion data. Specifically, if the emotion label of the market data is positive and the user's emotion is stable, the server generates a "buy" strategy, and if it is negative, the server generates a "sell" strategy.

[1406] Processing of investor offerings

[1407] The server then sends the generated trading strategy to the user's device and notifies the user. The user then accesses the trading platform based on the strategy provided and executes specific investment actions. In this process, the user is able to grasp market trends and risk factors in real time, enabling them to make timely and accurate investment decisions.

[1408] Specific examples

[1409] Consider the following example as a specific scenario.

[1410] 1. Data collection

[1411] For example, the server retrieves text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest, while simultaneously retrieving audio of a financial analyst from https: / / api.financial-audio.com / latest and image data showing rising stock prices from https: / / api.financial-images.com / latest.

[1412] 2. Collecting Emotional Data

[1413] The server collects the user's input text, speech, and data from biometric sensors (e.g., heart rate data). For example, if a user types, "What do you think about the market these days?", the server collects that text and speech.

[1414] 3. Data Analysis

[1415] For this collected text data, the server uses a generative AI model to obtain the emotion label "positive." After converting the voice data into text using the generative AI model, the server performs a similar emotion analysis and obtains the "positive" label. Furthermore, the generative AI model is used to analyze image data, obtaining the label "stock price rise."

[1416] 4. Emotion Data Analysis

[1417] The server uses an emotion engine to analyze the user's text, voice, and biometric sensor data, recognizing, for example, if the user indicates an emotion such as "excited."

[1418] 5. Generating a Trading Strategy

[1419] The server integrates all the analysis results and the user's emotional state, and generates a "buy" strategy since the overall results are positive and the user's emotions are stable.

[1420] 6. Offering to investors

[1421] The server sends a message to the user's terminal saying, "The current recommended trading strategy is 'BUY'." Based on this information, the user takes specific investment action through the trading platform.

[1422] This system comprehensively analyzes market data and user sentiment data, and generates and provides highly accurate trading strategies in real time, thereby significantly supporting investors' investment decisions.

[1423] The processing flow will be explained below.

[1424] Step 1:

[1425] The server retrieves the latest text data from the financial news API, specifically, collects JSON data of news articles from https: / / api.financial-news.com / latest.

[1426] Step 2:

[1427] The server collects audio data from the financial information service by downloading the audio file from https: / / api.financial-audio.com / latest and saving it.

[1428] Step 3:

[1429] The server retrieves financial image data from the image API by accessing the endpoint https: / / api.financial-images.com / latest and loading the retrieved image data as an image object using Image.open .

[1430] Step 4:

[1431] The server collects the user's input text, specifically, collects messages entered by the user in a chat window of the trading platform as a log.

[1432] Step 5:

[1433] The server collects voice input data from the user, specifically, stores voice data acquired through the voice assistant function.

[1434] Step 6:

[1435] The server collects the user's biometric sensor data, specifically recording heart rate, skin temperature, and other physiological data obtained from biometric sensors such as a smartwatch.

[1436] Step 7:

[1437] The server analyzes the collected text data using a sentiment analysis model. Specifically, it runs a text sentiment analysis pipeline using the DistilBERT model and outputs sentiment labels (positive, neutral, or negative) for news articles.

[1438] Step 8:

[1439] The server uses a model to convert audio data into text. Specifically, it transcribes the audio data using the Wav2Vec2 model and stores the resulting text data.

[1440] Step 9:

[1441] The server analyzes the text extracted from the audio data using a sentiment analysis model. Specifically, it uses the DistilBERT model to obtain sentiment labels (positive, neutral, negative) for the previously saved text data.

[1442] Step 10:

[1443] The server uses an image classification model to analyze the image data. Specifically, it uses a Vision Transformer (ViT) model to obtain labels for the information the image represents (e.g., a graph of rising stock prices).

[1444] Step 11:

[1445] The server uses an emotion engine to analyze the user's text, voice, and biometric sensor data, inferring the user's emotional state and obtaining emotion labels.

[1446] Step 12:

[1447] The server integrates all data analysis results, specifically, compiles emotion and image labels obtained from each data source into a single list and performs a comprehensive evaluation.

[1448] Step 13:

[1449] The server automatically generates a trading strategy based on the integrated analysis results and user sentiment data. Specifically, if the sentiment analysis results and user sentiment are stable, it generates a "BUY" strategy, and if not, it generates a "SELL" strategy.

[1450] Step 14:

[1451] The server sends the generated trading strategy to the user's terminal, notifying the user by displaying a message saying, "The current recommended trading strategy is 'BUY'."

[1452] Step 15:

[1453] The user accesses the trading platform based on the provided trading strategy and takes specific investment actions, such as "buying" or "selling" according to the displayed strategy.

[1454] In this way, the server can comprehensively analyze market data and user sentiment data, and generate and provide highly accurate trading strategies in real time, thereby significantly supporting users' investment decisions.

[1455] Example 2

[1456] 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."

[1457] It is extremely difficult for investors to make instant decisions amidst the rapid fluctuations in information and vast amounts of data in financial markets. Furthermore, because investors' emotional state has a significant impact on investment decisions, strategies that ignore emotions have limited effectiveness. Therefore, in order for investors to make faster and more accurate decisions, a system is needed that can comprehensively analyze market data and emotional data and provide investors with specific trading strategies in real time.

[1458] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1459] In this invention, the server includes means for collecting financial market data, means for performing multimodal analysis of the collected data and user emotion data, means for generating a trading strategy based on the analysis results and the user emotion data, and means for providing the generated trading strategy to investors, thereby enabling investors to grasp market trends and obtain highly accurate trading strategies that take their own emotional state into account in real time.

[1460] "Financial market data" refers to various information obtained from financial markets, including stock prices, exchange rates, interest rates, economic indicators, news articles, reports, etc.

[1461] "Multimodal analysis" is a method for integrating and analyzing data of different formats (text, audio, images, biometric sensors, etc.).

[1462] A "trading strategy" refers to decisions and plans for buying and selling in financial markets, and specifically includes recommendations for "buying" and "selling."

[1463] "Emotion data" is data that represents the psychological state of the user, and includes information obtained from text messages, voice input, biometric sensors, and the like.

[1464] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to analyze data and generate a specific outcome (e.g., a sentiment label or a trading strategy).

[1465] A "prompt sentence" is an input sentence that causes a generative AI model to perform a specific analysis or generation task.

[1466] This invention relates to a system for supporting investment analysis in financial markets, specifically an "Intelligent Market Sensor" system. This system uses generative AI models and AGI to comprehensively analyze multimodal data, including text, voice, images, and biometric sensors, to detect market trends, risk factors, and investment opportunities in real time, and provide specific trading strategies to investors.

[1467] Data collection

[1468] The server retrieves text data from a financial news API and collects audio data from a financial information service. It also retrieves image data from an image API and subjects these data to an analysis step. For example, the server may retrieve text data such as "The market is currently trending upward" from https: / / api.financial-news.com / latest. At the same time, it retrieves audio of a financial analyst's speech from https: / / api.financial-audio.com / latest and image data showing rising stock prices from https: / / api.financial-images.com / latest.

[1469] Emotional Data Collection

[1470] The server collects emotion-related data from the user's text messages, speech input, and biometric sensors. This data includes, for example, the user's text input (e.g., "What do you think about the market these days?"), speech, and biometric data such as heart rate.

[1471] Data analysis

[1472] The server analyzes the collected text data using a generative AI model (for example, OpenAI's GPT-4) to obtain an emotion label. It also analyzes voice data in the same way, converts it into text, and then obtains an emotion label. This conversion is performed using Google's Speech-to-Text API. Image data is analyzed using an image classification model such as Google's Vision API, and the content is classified as a label. For example, text data such as "The market is currently on the rise" is input into a generative AI model to create a prompt that obtains the emotion label "positive."

[1473] Emotional Data Analysis

[1474] The server uses an emotion engine to analyze data obtained from the user's text, voice, and biometric sensors to recognize the user's emotional state. For example, it identifies emotions such as "excited" from the user's heart rate and facial expression data. This recognized emotion data is used in the subsequent trading strategy generation step.

[1475] Trading Strategy Generation

[1476] The server automatically generates a trading strategy based on the analysis results and the user's emotional data. Specifically, if the market data's emotional label is positive and the user's emotional state is stable, a "buy" strategy is generated; if it is negative, a "sell" strategy is generated. The server prompts the generating AI model with the following prompt: "The current market data is positive. The user's emotional state is also stable. What trading strategy do you recommend?"

[1477] Offering to investors

[1478] The server then sends the generated trading strategy to the user's device and notifies the user. Based on the strategy, the user can access the trading platform and take specific investment actions. This process allows users to grasp market trends and risk factors in real time and make timely and accurate investment decisions.

[1479] This system comprehensively analyzes market data and user sentiment data, and generates and provides highly accurate trading strategies in real time, thereby significantly supporting investors' investment decisions.

[1480] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1481] Step 1:

[1482] Data collection

[1483] The server first obtains text data from a financial news API (e.g., https: / / api.financial-news.com / latest). It receives the response data (news article text) from the API as input and stores the text data in an internal database as output. At the same time, it collects audio data from a financial information service and obtains the audio file using an audio API (e.g., https: / / api.financial-audio.com / latest). It also obtains image data from an image API (e.g., https: / / api.financial-images.com / latest) and stores them in an internal database. Specifically, the server sends an HTTPS request and parses the JSON-formatted response to extract the necessary information.

[1484] Step 2:

[1485] Emotional Data Collection

[1486] The server collects emotion-related data from text messages entered by the user, audio from conversations, and biometric sensors. As input, it receives text messages, audio files, and biometric sensor data sent from the user's device. As output, it generates structured data formatted for analysis. For example, if a user types, "What do you think about the market these days?", the server collects the text and also captures the audio data and saves it as an audio file.

[1487] Step 3:

[1488] Data analysis

[1489] The server inputs the collected text data into a generative AI model (e.g., OpenAI's GPT-4) to obtain an emotion label. The collected text data is provided as input, and a generated emotion label (e.g., "positive" or "negative") is obtained as output. Voice data is also converted into text using voice recognition software (e.g., Google's Speech-to-Text API), which is then analyzed by the generative AI model. Image data is analyzed using an image classification model (e.g., Google's Vision API), and the content of the image is classified as a label. Specifically, the analysis results in label information such as "the market is on the rise," "financial analysts' comments are positive," and "image showing rising stock prices."

[1490] Step 4:

[1491] Emotional Data Analysis

[1492] The server uses an emotion engine to analyze data obtained from the user's text, voice, and biometric sensors to recognize the user's emotional state. The server provides the user's text data, voice data, and biometric data as input, and identifies the user's emotional state (e.g., "excited" or "calm") as output. Specifically, the server runs the text analysis engine, voice analysis engine, and biometric sensor analysis engine to evaluate the user's overall emotional state.

[1493] Step 5:

[1494] Trading Strategy Generation

[1495] The server automatically generates trading strategies based on market data and user sentiment data. It provides analysis results and sentiment data as input, and generates specific trading strategies such as "buy" or "sell" as output. This is done by inputting prompts such as "Current market data is positive. The user's sentiment state is stable. What trading strategy do you recommend?" into the generative AI model, and obtaining output from the model. For example, if the overall market data is positive, a "buy" strategy is generated.

[1496] Step 6:

[1497] Offering to investors

[1498] The server sends the generated trading strategy to the user's device and notifies the user. It provides the generated trading strategy data as input and sends a notification in the form of a message to the user's device as output. The user receives the notification and accesses a trading platform (e.g., an online securities site or a dedicated app) to perform specific investment actions. Specifically, the server sends a notification to the user's device via a REST API, and the user can make investments through the trading application.

[1499] (Application example 2)

[1500] 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."

[1501] Conventional trading strategy generation systems are limited to analyzing financial market data and do not address the optimization of sales strategies in brick-and-mortar stores. Furthermore, because strategies are not generated taking into account the emotional state of customers, it is difficult to improve customer satisfaction and sales efficiency. Therefore, there is a need for a method to optimize brick-and-mortar store sales strategies in real time, reflecting the emotional state of customers.

[1502] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1503] In this invention, the server includes means for collecting financial market data and data from a physical store, means for performing multimodal analysis of the collected data and recognizing customer emotion data, means for generating trading and sales strategies based on the analysis results and the customer emotion data, and means for providing the generated trading and sales strategies. This makes it possible to generate sales strategies based on product trends and customer emotions in the physical store in addition to analyzing financial market data, thereby improving customer satisfaction and maximizing sales.

[1504] "Financial market data" refers to information such as text, audio, images, and video related to financial markets, including stock prices, trading volume, corporate financial status, economic indicators, and news.

[1505] "In-store data" refers to information such as customer behavior, purchase history, inventory status, and promotional effectiveness in physical stores, and includes data obtained from IoT sensors and cameras.

[1506] "Multimodal analysis" is a technology that integrates and analyzes multiple data formats, such as text, audio, images, and video, and is a method of extracting useful information from data using generative AI models.

[1507] "Customer emotion data" refers to the emotional state analyzed based on data obtained from the customer's facial expressions, voice, text messages, and biometric sensors, and includes emotional categories such as positive, negative, excited, and calm.

[1508] A "trading strategy" is a plan that outlines the method or direction for executing transactions in financial markets, including buying and selling decisions.

[1509] A "sales strategy" is a plan that outlines the methods and policies for selling products in physical stores, including strengthening promotions and adjusting inventory.

[1510] A "generative AI model" is a model that uses artificial intelligence techniques to analyze collected data and is trained to perform specific tasks, such as pattern recognition and predictions in the data.

[1511] "Emotion engine" refers to technology that recognizes and labels emotions from a user's text, voice, and biometric sensors, and is used to perform data analysis based on their emotional state.

[1512] "Increased promotion" refers to increasing sales promotion activities such as advertising and discounts for specific products or services in order to promote product sales.

[1513] An "IoT sensor" is a sensor connected to the Internet that measures and collects physical environmental data and transmits the data in real time.

[1514] "Sales maximization" refers to management and strategic activities aimed at increasing the profitability of a store or business as much as possible.

[1515] The present invention is a system that collects and analyzes data from financial markets and data from physical stores, and provides users with optimal trading and sales strategies. Specific embodiments for carrying out the present invention will be described below.

[1516] System Configuration

[1517] Hardware and Software

[1518] 1. Server: A central processing unit that collects data, analyzes it, and generates strategies. It has a high-performance processor and sufficient storage.

[1519] 2. IoT Sensors: These sensors collect environmental data within physical stores, measuring temperature, humidity, movement, etc.

[1520] 3. Camera: Captures customer movements and facial expressions in the physical store and sends them to the server as image data.

[1521] 4. Voice Assistant: A device that conducts voice interactions with customers and transmits voice data to a server.

[1522] 5. Generative AI models: Machine learning models that analyze text, voice, and image data to generate investment and sales strategies.

[1523] 6. Emotion engine: Software technology for analyzing the user's emotional state and obtaining emotion labels.

[1524] Data collection methods

[1525] The server collects financial market data and data from physical stores. Financial market data includes text data, audio data, image data, and video data obtained from news APIs and financial information providers. Physical store data is collected from IoT sensors, cameras, and voice assistants.

[1526] Data Analysis Methods

[1527] The server uses generative AI models to perform multimodal analysis of the collected data: text data is analyzed using natural language processing (NLP) techniques, voice data is converted to text and then labeled with an emotion label, and image data is analyzed using image classification models to recognize the customer's emotional state.

[1528] Emotion data analysis method

[1529] The server uses an emotion engine to analyze data from users' text, voice, and biometric sensors to recognize their emotional state. For example, if a customer is expressing positive emotions, it can provide strategies based on that.

[1530] Strategy Generation Method

[1531] Based on the analysis results and customer sentiment data, the server generates trading and sales strategies, for example, if the financial market data is positive and the customer sentiment state is positive, it will suggest a "buy" trading strategy or strengthen the promotion of a specific product.

[1532] Strategy delivery methods

[1533] The server provides the generated trading and sales strategies to the users who will actually use them, and the users then take specific investment and sales actions based on these strategies.

[1534] Specific examples

[1535] For example, the server retrieves market text data from the URL "https: / / api.financial-news.com / latest" and assigns an emotion label of "positive." It also analyzes customer facial expressions through cameras in physical stores to determine whether the customer is enjoying themselves. Based on this data, the server generates a message saying, "The current recommended trading strategy is 'BUY'," and sends it to the user's device. It also suggests to the store operator to strengthen promotions of specific products.

[1536] Example prompt sentence:

[1537] "Analyze recent customer sentiment and correlate it with point-of-sale data to optimize your sales strategy."

[1538] "How can we update the product's availability and increase promotions when we recognize high customer satisfaction for a particular product?"

[1539] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1540] Step 1:

[1541] Data collection

[1542] The server collects financial market data from news APIs, financial information services, image APIs, etc. For example, it obtains text data from "https: / / api.financial-news.com / latest," audio data from "https: / / api.financial-audio.com / latest," and image data from "https: / / api.financial-images.com / latest." It also simultaneously collects data from IoT sensors, cameras, and voice assistants in physical stores. Various APIs are used as input, and the collected text, audio, and image data are obtained as output.

[1543] Step 2:

[1544] Data Preprocessing

[1545] The server converts the collected raw data into a format suitable for analysis: it standardizes text data, converts voice data into text using voice recognition technology, and preprocesses image data to standardize pixel information. It receives the collected text, voice, and image data as input, and outputs a standardized dataset.

[1546] Step 3:

[1547] Emotional Data Collection

[1548] The server collects user input text, voice, and biometric sensor data. Specifically, voice data is obtained through a voice assistant, and biometric sensor data is collected from heart rate and facial expression analysis. The input is text, voice, and biometric sensor data from the user, and the output is data that can be labeled with emotions.

[1549] Step 4:

[1550] Data analysis

[1551] The server uses a generative AI model to analyze collected text, audio, and image data to obtain emotion labels. For example, it performs natural language processing (NLP) on text data, emotion recognition on audio data, and content classification on image data. It receives a standardized dataset as input and obtains emotion labels as output.

[1552] Step 5:

[1553] Emotional Data Analysis

[1554] The server uses an emotion engine to comprehensively analyze the user's emotional data, recognize the user's emotional state from text, voice, and biometric sensor data, and generate an emotion label based on this. It receives the data collected from the user as input and obtains the user's emotional state data as output.

[1555] Step 6:

[1556] Strategy Generation

[1557] The server generates trading and sales strategies based on the analysis results and user emotion data, for example, making buying and selling decisions or proposing promotion enhancements. It receives emotion labels and analysis results as input and obtains specific strategies as output.

[1558] Step 7:

[1559] Strategy Providing

[1560] The server provides the generated trading and selling strategies to the user. Specifically, it sends a message to the user's terminal saying "The current recommended trading strategy is 'BUY'." It receives the generated strategies as input and displays a notification message on the user's terminal as output.

[1561] 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.

[1562] 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.

[1563] 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.

[1564] 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.

[1565] 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.

[1566] 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.

[1567] 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).

[1568] 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.

[1569] 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."

[1570] 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.

[1571] 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).

[1572] 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.

[1573] 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.

[1574] 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.

[1575] 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.

[1576] 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.

[1577] 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.

[1578] 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.

[1579] 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.

[1580] 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.

[1581] 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.

[1582] The following is further disclosed regarding the above embodiment.

[1583] (Claim 1)

[1584] a means of collecting financial market data;

[1585] A means of multimodal analysis of the collected data;

[1586] means for generating a trading strategy based on the analysis results;

[1587] A means for providing the generated trading strategies to investors;

[1588] A system including:

[1589] (Claim 2)

[1590] 10. The system of claim 1, wherein the system collects at least one of text data, audio data, image data, and video data.

[1591] (Claim 3)

[1592] 10. The system of claim 1, wherein the generated trading strategy includes buy or sell recommendations based on the collected and analyzed data.

[1593] "Example 1"

[1594] (Claim 1)

[1595] means of collecting financial market information;

[1596] A means of multimodal analysis of the collected information;

[1597] A means for analyzing text, voice, and image data using a generative AI model;

[1598] means for generating a trading strategy based on the analysis results;

[1599] A means for providing the generated trading strategies to investors;

[1600] A system including:

[1601] (Claim 2)

[1602] 10. The system of claim 1, wherein the system collects at least one of text data, audio data, and image data.

[1603] (Claim 3)

[1604] 10. The system of claim 1, wherein the system performs sentiment analysis using a generative AI model.

[1605] (Claim 4)

[1606] 10. The system of claim 1, wherein the generated trading strategy includes buy or sell recommendations based on the collected and analyzed data.

[1607] "Application Example 1"

[1608] (Claim 1)

[1609] a means of collecting financial market data;

[1610] A means of multimodal analysis of the collected data;

[1611] means for generating a trading strategy based on the analysis results;

[1612] A means for providing the generated trading strategies to investors;

[1613] means for collecting user activity data;

[1614] A means for analyzing collected user activity data and generating optimal content;

[1615] means for providing the generated content to a user;

[1616] A system including:

[1617] (Claim 2)

[1618] 10. The system of claim 1, wherein the system collects at least one of text data, audio data, image data, and video data.

[1619] (Claim 3)

[1620] 10. The system of claim 1, wherein the generated trading strategies and content include recommendations based on the collected and analyzed data.

[1621] "Example 2: Combining Emotion Engines"

[1622] (Claim 1)

[1623] a means of collecting financial market data;

[1624] a means for multimodal analysis of the collected data and the user's emotion data;

[1625] means for generating a trading strategy based on the analysis results and the user sentiment data;

[1626] A means for providing the generated trading strategies to investors;

[1627] A system including:

[1628] (Claim 2)

[1629] 10. The system of claim 1, wherein the system collects at least one of text data, audio data, image data, and biometric sensor data.

[1630] (Claim 3)

[1631] 10. The system of claim 1, wherein the generated trading strategy includes buy or sell recommendations based on the collected and analyzed data and user sentiment data.

[1632] "Application example 2 when combining emotion engines"

[1633] (Claim 1)

[1634] a means of collecting financial market data and brick-and-mortar data;

[1635] A means of multimodal analysis of collected data and recognizing customer emotion data;

[1636] means for generating trading and sales strategies based on the analytical results and customer sentiment data;

[1637] a means for providing the generated trading and sales strategies;

[1638] A system including:

[1639] (Claim 2)

[1640] 10. The system of claim 1, wherein the system collects at least one of text data, audio data, image data, and video data.

[1641] (Claim 3)

[1642] 10. The system of claim 1, wherein the generated trading and selling strategies include buy or sell recommendations and promotional enhancements based on the collected and analyzed data. [Explanation of symbols]

[1643] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means of collecting financial market data; A means of multimodal analysis of the collected data; means for generating a trading strategy based on the analysis results; A means for providing the generated trading strategies to investors; A system including:

2. 10. The system of claim 1, wherein at least one of text data, audio data, image data, and video data is collected.

3. 10. The system of claim 1, wherein the generated trading strategy includes buy or sell recommendations based on the collected and analyzed data.

Images