System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting information; means for formatting the collected information and storing the formatted information in a database; means for extracting features from the stored information and training a AI model; means for predicting a future trend using the trained AI model; means for transmitting a prediction result to a user interface and converting the prediction result into a displayable format; and means for obtaining feedback from a user and retraining the AI model.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] Currently, accurate, real-time understanding of market trends is essential for mid- to large-scale companies to make effective strategic decisions. However, there is no integrated system that collects information from vast data sources, formats the data, builds and operates predictive models, and incorporates user feedback. As a result, companies are often unable to respond quickly to market changes, making it difficult to make optimal decisions. Another issue is the lack of a way to efficiently incorporate the insights of users and experts with specialized knowledge and improve the accuracy of their predictions. [Means for solving the problem]

[0005] To solve the above problems, the present invention provides the following means. First, a data collection means acquires data from multiple data sources in real time and converts it into multiple data frames. Next, these data frames are combined and the formatted data is stored in a database. Next, a means is provided for extracting features from the stored data and training an AI model. The present invention also implements a means for predicting future trends using the trained AI model, sending the prediction results to a user interface, and converting them into a displayable format. Furthermore, the present invention includes a means for receiving predictions and opinions entered by users through a form, storing them in a database, and retraining the AI ​​model using the stored feedback. This makes it possible to incorporate user and expert insights, improve prediction accuracy, grasp market trends in real time, and support effective strategic decision-making.

[0006] "Data collection means" refers to a method or device for acquiring data from multiple data sources in real time.

[0007] "Database" refers to a data storage system for storing and managing formatted data.

[0008] "Features" refer to the data attributes and variables extracted from training data that are necessary for learning a predictive model.

[0009] "AI model" refers to an algorithm or mathematical model that uses artificial intelligence technology to analyze data and predict future trends.

[0010] "Training" refers to the process of using features to teach an AI model and improve its accuracy.

[0011] "Prediction means" refers to a method or device that uses a trained AI model to estimate future trends.

[0012] "User interface" refers to a screen or device that visually displays prediction results and allows users to easily understand the information.

[0013] "Feedback" refers to opinions and reactions based on thoughts and predictions provided by users.

[0014] "Retraining" refers to the process of retraining an AI model based on the feedback it receives to improve the accuracy of its predictions.

[0015] An "expert" is an individual or group with advanced knowledge or experience in a particular field who can provide insights and analytical reports.

[0016] "Insights" refers to analytical information based on deep understanding and opinions provided by experts. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0025] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0038] A system for implementing the present invention combines a multi-step process that includes data collection, data analysis and trend prediction, receiving and processing user feedback, integrating expert insights, and presenting the predictive results.

[0039] Data collection

[0040] The server first collects data in real time. To do this, it obtains data from multiple data sources, such as news feeds, social media, and economic indicators, via APIs. The obtained data is then formatted and converted into a data frame. Multiple data frames are then combined to create a unified dataset. This unified dataset is then stored in a database. This process requires efficient data collection and storage.

[0041] Data analysis and trend forecasting

[0042] The server then reads the stored data from the database and extracts the necessary features. It then trains the data using an AI model, which uses algorithms such as random forest and LSTM. Once trained, the model is used to predict future trends. The prediction results are then formatted so that they are displayed to the user in a visually understandable format and sent to the user's device.

[0043] Receiving and processing feedback

[0044] Users access the system through their devices and input their predictions and opinions. For example, they can enter a prediction such as, "I think sales will increase next quarter." These predictions and opinions are sent to the server, which stores the received feedback in a database and further uses this feedback to update the training data and retrain the AI ​​model.

[0045] Integrating expert insights

[0046] Experts access the platform and input their own insights and analysis reports. These insights are sent to the server and stored in a database. The AI ​​model is then retrained based on the stored insights to improve its prediction accuracy. This process allows expert knowledge to be incorporated into the model.

[0047] Prediction results

[0048] When a user accesses the terminal, it sends a request to the server to obtain the latest trend forecast information. The server returns the latest forecast results, and the terminal displays these results on the dashboard in the form of interactive graphs and charts, allowing users to grasp the latest market trends at a glance.

[0049] As described above, the present invention provides a system that consistently performs processes from data collection to prediction and feedback integration, thereby enabling companies to make more effective and faster decisions and supporting them in responding quickly to market changes.

[0050] The processing flow will be explained below.

[0051] Step 1:

[0052] Data collection and formatting (server)

[0053] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. The data is then formatted and converted into a data frame. These data frames are then combined to create a unified dataset. Finally, this unified dataset is stored in a database.

[0054] Step 2:

[0055] Feature extraction (server)

[0056] The server reads the stored data from the database and extracts the necessary features. Features are the data attributes and variables required to train a predictive model. These are important elements for correctly forecasting market trends.

[0057] Step 3:

[0058] AI model training (server)

[0059] The server trains an AI model based on the extracted features. Models used here include random forests and LSTM. The server uses these algorithms to learn the training data. Cross-validation and other methods are performed to improve the model's performance.

[0060] Step 4:

[0061] Future Trend Prediction (Server)

[0062] The server uses the trained AI model to predict future trends, and the prediction results are displayed in a visually easy-to-understand format (e.g., graphs and charts), allowing users to understand the prediction results at a glance.

[0063] Step 5:

[0064] Presenting prediction results (device)

[0065] When a user accesses the terminal, it sends a request to the server to obtain the latest trend forecast information. The server returns the forecast results to the terminal, which then displays them on the dashboard in the form of interactive graphs and charts, allowing users to understand market trends in real time.

[0066] Step 6:

[0067] Receiving user feedback (device, user)

[0068] Users input their predictions and opinions through their own devices. For example, they input predictions such as, "I think sales will increase in the next quarter." This feedback is sent to the server through a form.

[0069] Step 7:

[0070] Feedback storage and processing (server)

[0071] The server receives the feedback sent by the user and stores it in a database, after which the AI ​​model is retrained based on the stored feedback, a process that allows the accuracy of the prediction model to be improved.

[0072] Step 8:

[0073] Integration of expert insights (server, user)

[0074] Experts access the platform and input their own insights and analysis reports. The server stores these insights in a database and uses them to retrain the AI ​​model. By incorporating the experts' knowledge into the model, the accuracy of predictions can be further improved.

[0075] The above is the specific processing flow of this system.

[0076] Example 1

[0077] 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."

[0078] Traditional data analysis systems struggled to consistently collect data, forecast trends, and integrate feedback. This made it difficult for companies to obtain the latest market trend information in a timely manner to make quick decisions. Furthermore, there was a lack of a way to effectively integrate feedback and insights from users and experts to improve forecast accuracy.

[0079] 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.

[0080] In this invention, the server includes a data collection means, a means for formatting the collected data and storing it in a database, a means for extracting features from the stored data and training a machine learning model, a means for predicting future trends using the trained machine learning model, a means for sending the prediction results to a user interface and converting them into a displayable format, a means for obtaining feedback from users and retraining the machine learning model, a means for receiving expert insights through the platform and storing them in a database, and a means for retraining the machine learning model based on the stored expert insights to improve prediction accuracy. This enables companies to collect data in real time, make highly accurate trend predictions, and build a feedback loop that reflects the opinions of users and experts.

[0081] "Data collection means" refers to the means for obtaining data in real time from various data sources such as news feeds, social media, and economic indicators.

[0082] "Format" is the process of converting collected data into a usable format and carrying out the data cleaning and formatting required for storing it in a database.

[0083] "Means for storing in a database" refers to the means for storing formatted data in a database so that it can be efficiently searched and retrieved.

[0084] "Features" are attributes and values ​​of data extracted and transformed to be fed into a machine learning model, and contain information that is important for analysis and prediction.

[0085] A "machine learning model" is an algorithm or model that is trained using data to make predictions or classifications based on future data.

[0086] "Training means" refers to the means for training a machine learning model using collected data to improve the accuracy of the model.

[0087] A "means for predicting future trends" is a means for predicting trends in future data or events using a trained machine learning model.

[0088] A "user interface" is an interface through which a user accesses a system to input information or obtain information from the system.

[0089] The "means for converting into a displayable format" refers to a means for converting the prediction results into a visually easy-to-understand format (for example, a graph or chart) and displaying it on a user interface.

[0090] "Feedback" refers to predictions, opinions, and insights provided by users and experts.

[0091] "Means for obtaining feedback" refers to the means for collecting and storing feedback from users and experts.

[0092] "Retraining" refers to the process of retraining a machine learning model using feedback and new data to improve the model's accuracy.

[0093] "Platform" means a designated environment or system for experts to input and integrate their insights and analytical reports into the system.

[0094] "Insights" refers to detailed analysis and opinions provided by experts, and are important information for improving forecast accuracy.

[0095] The system of the present invention combines a multi-step process, including data collection, data analysis and trend prediction, receiving and processing user feedback, integrating expert insights, and presenting the prediction results. As a specific implementation method of the present invention, the following steps are described in detail.

[0096] Data collection

[0097] The server collects data in real time from multiple data sources, such as news feeds, social media, and economic indicators, via APIs. For example, the server uses Twitter's API to collect tweets based on specific keywords. It converts this data into a data frame using the Pandas library. The server then combines multiple data frames to create a unified dataset, which is then stored in an SQL database.

[0098] Data analysis and trend forecasting

[0099] The server reads the stored data from the database and extracts features. For example, the server extracts the number of tweets in the past 24 hours and the percentage of tweets with positive context. The server then trains a machine learning model such as random forest or LSTM using Scikit-learn or TensorFlow. The trained model is used to predict future trends. The prediction results are formatted into a visually easy-to-understand format (e.g., graphs or charts) using the Plotly library and sent to the user's device.

[0100] Receiving and processing feedback

[0101] Users use their own devices to input their predictions and opinions into the system. For example, a user can input a prediction such as, "I think sales will increase next quarter." The device then sends the input feedback to the server. The server stores the received feedback in a database and updates the training data based on the stored feedback. The server then retrains the machine learning model using the updated data.

[0102] Integrating expert insights

[0103] Experts access the platform and input their own insights and analysis reports. The device then sends these insights to the server, which stores them in a database and retrains the machine learning model based on the stored insights. This incorporates expert knowledge into the model, improving its prediction accuracy.

[0104] Prediction results

[0105] The user's device sends a request to the server to get the latest trend forecast information, and the server returns the latest forecast results, which the device displays on a dashboard in the form of interactive graphs and charts.

[0106] Specific examples

[0107] For example, a user can enter the following prompt into the system:

[0108] "Collect the latest tweet data and make supply and demand forecasts."

[0109] "Create a sales forecast for the next quarter and display it in a chart."

[0110] "Update the AI ​​model based on user feedback."

[0111] In response to these prompts, the server collects, analyzes, and processes appropriate data. Furthermore, by incorporating expert insights into the model, forecast accuracy can be improved. In this way, the system of the present invention helps companies respond quickly to market changes and make effective decisions.

[0112] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0113] Step 1: Data collection

[0114] The server uses APIs to collect data in real time from various data sources, such as news feeds, social media, and economic indicators. For example, the Twitter API is used to collect tweets based on specific keywords. The server periodically executes scheduled tasks to obtain the data. The collected data is converted into a data frame using the Pandas library. The input is raw data from each data source, and the output is a formatted data frame.

[0115] Step 2: Format and save the data

[0116] The server reformats the collected data using the Pandas library. This reformatting process includes removing duplicate data, handling missing values, and normalizing text. The reformatted data is then combined to create a unified dataset. This dataset is then stored in a SQL database. The input is an unformed data frame, and the output is the reformatted and combined dataset.

[0117] Step 3: Feature extraction

[0118] The server reads the stored data from the database and extracts features, such as the number of tweets per hour or the percentage of tweets with positive context. The Scikit-learn library is used to encode these features. The input is the dataset, and the output is a list of numerically encoded features.

[0119] Step 4: Train the model

[0120] The server trains the extracted features using machine learning models such as random forests and LSTM. For example, it uses past tweet data to predict future trends. Libraries such as TensorFlow and Scikit-learn are used for this training. The accuracy of the trained model is verified and saved. The input is a list of features, and the output is a trained machine learning model.

[0121] Step 5: Predicting trends

[0122] The server uses the trained model to predict future trends, for example, forecasting sales for the next quarter. The prediction results are visualized using the Plotly library and output in a format that is easy for users to understand. The input is the machine learning model and features, and the output is the visualized prediction results.

[0123] Step 6: Receiving feedback

[0124] Users input their predictions and opinions into the system through their terminals. For example, they input a prediction such as, "I think sales will increase next quarter." The terminals collect this feedback and send it to the server. The input is the user's feedback, and the output is the feedback data sent to the server.

[0125] Step 7: Process feedback and retrain

[0126] The server stores the received feedback in a database and updates the training data based on the stored feedback. The AI ​​model is retrained using the updated data, which improves the model's accuracy. The input is the feedback data, and the output is an updated machine learning model.

[0127] Step 8: Integrating expert insights

[0128] Experts access the platform and input their insights and analysis reports. The device sends them to the server, which stores the received insights in a database. The stored insights are used to retrain the machine learning model, further improving its prediction accuracy. The input is the expert's insights, and the output is the retrained machine learning model.

[0129] Step 9: Presenting the prediction results

[0130] The user's device requests the latest forecast results from the server. The server returns the latest forecast results, and the device displays these results on a dashboard as interactive graphs and charts, allowing the user to understand the latest market trends at a glance. The input is a request for forecast results, and the output is the forecast results in a displayable format.

[0131] (Application example 1)

[0132] 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."

[0133] Conventional ad delivery systems lacked real-time market trend analysis, making effective targeting and personalization difficult. They also struggled to quickly incorporate user and expert feedback, delaying the optimization of ad campaigns. Furthermore, limited means of collecting and analyzing user behavior data made it difficult to accurately evaluate advertising effectiveness.

[0134] 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.

[0135] In this invention, the server includes a data collection means, a means for formatting the collected data and storing it in a database, a means for extracting features from the stored data and training an AI model, a means for predicting future trends using the trained AI model, a means for sending the prediction results to a user interface and converting them into a displayable format, a means for receiving feedback from users and retraining the AI ​​model, a means for receiving insights and analysis reports from experts and storing them in a database and retraining the AI ​​model, and a means for generating personalized advertisements based on the collected and analyzed data and displaying them on user terminals. This enables effective targeting based on real-time market trend analysis, optimization of advertising campaigns that quickly reflect user and expert feedback, and accurate evaluation of advertising effectiveness by collecting and analyzing user behavior data.

[0136] "Data collection means" refers to a means for obtaining data in real time from multiple data sources such as news feeds, social media, and economic indicators.

[0137] "Database" means a centralized data repository for storing uniform data sets for later use in analysis and prediction.

[0138] A "feature extraction method" is a method for identifying information necessary for trend prediction from collected data and converting it into a format that can be used to train an AI model.

[0139] "AI model training method" refers to a method of training an AI model (such as random forest or LSTM) using extracted features to enable it to predict future trends.

[0140] A "trend forecasting tool" is a tool for forecasting future trends using a trained AI model.

[0141] The "user interface transmission means" is a means for converting the prediction results into a format that is easy for the user to understand and transmitting it to the user's terminal.

[0142] "Feedback acquisition means" refers to the means of collecting predictions and opinions from users and reflecting them in the system.

[0143] "AI model retraining method" refers to a method of retraining an AI model based on feedback from users and experts to improve its accuracy.

[0144] The "means for receiving expert insights" is a means for receiving insights and analysis reports from experts, storing them in a database, and then retraining the AI ​​model.

[0145] "Personalized advertisement generation means" refers to a means for generating advertisements optimized for individual users based on collected and analyzed data and displaying them on the user's terminal.

[0146] The system that realizes this application example comprises the smartphone app "AdTrend." The server collects data in real time from multiple data sources, such as news feeds, social media, and economic indicators. The collected data is formatted and converted into a data frame, and these data frames are then combined to create a unified dataset. This process uses Python's Pandas framework.

[0147] Next, the server reads the stored data from the database and extracts the necessary features. Based on the extracted features, it trains an AI model using algorithms such as random forest and LSTM. This training generates an AI model that can predict future trends. The model is trained using the Scikit-learn library and generates prediction results.

[0148] The server then formats these forecasts in JSON format and sends them to the user's smartphone, where they are displayed as interactive charts and graphs. This dashboard helps users visually understand trends.

[0149] Users can enter their predictions and opinions into the system through a form. This feedback is sent to the server and stored in a database. The stored feedback is used as new training data to retrain the AI ​​model. The Flask framework is used to receive and process user feedback.

[0150] In addition, insights and analysis reports from experts are also sent to the server and stored in the database, allowing the experts' knowledge to be utilized to improve the prediction accuracy of the AI ​​model. In this process, the experts' insights are incorporated into the retraining of the AI ​​model.

[0151] Regarding the generation of personalized ads, the collected data is used to generate ads optimized for individual user attributes. These ads are delivered and displayed on the user's device. User behavior data is collected and analyzed to evaluate the effectiveness of the ads. Ad campaigns are optimized using Python and related libraries.

[0152] Examples of specific prompts include:

[0153] Trend Forecast:

[0154] Analyze the characteristics of the most popular advertising campaigns in the next quarter and propose effective advertising strategies.

[0155] Feedback received:

[0156] We collect user opinions about the ads currently being served and use that data to personalize ads.

[0157] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0158] Step 1:

[0159] The server collects data in real time from multiple data sources, such as news feeds, social media, and economic indicators. It uses an API to retrieve the data and converts it into a Pandas data frame. The input is raw data from each data source, and the output is organized data in a data frame format.

[0160] Step 2:

[0161] The server combines the multiple formatted data frames to create a unified dataset and stores it in the database. The input is multiple data frames, and the output is the combined unified dataset. Specifically, the server combines the data frames using functions such as the merge function in Pandas and stores it in the database.

[0162] Step 3:

[0163] The server reads the stored data from the database and extracts the necessary features. The input is a unified dataset, and the output is the features required for training the AI ​​model. Specifically, it uses Pandas and Scikit-learn to extract the features and prepare them as training data.

[0164] Step 4:

[0165] The server uses the extracted features to train an AI model. The input is the features, and the output is a trained AI model (e.g., random forest or LSTM). Specifically, the server uses the Scikit-learn library to train the model and check its prediction accuracy.

[0166] Step 5:

[0167] The server uses a trained AI model to predict future trends. The input is newly collected data, and the output is the future trend prediction result. Specifically, new data is input into the model and a prediction result is generated.

[0168] Step 6:

[0169] The server formats the prediction results in JSON format and sends them to the user's device. The input is the prediction result data, and the output is the JSON-formatted prediction result sent to the user's device. Specifically, Flask is used to communicate data between the server and the device.

[0170] Step 7:

[0171] The terminal displays the received prediction results as interactive charts and graphs. The input is the prediction result data in JSON format, and the output is the prediction result displayed graphically. Specifically, a dashboard is created using a JavaScript library (e.g., D3.js) to provide visual feedback to the user.

[0172] Step 8:

[0173] Users enter their predictions and opinions into the system through a form. The input is the user's feedback, and the output is the feedback data sent to the server. Specifically, the system uses a combination of HTML forms and JavaScript to send user input to the server.

[0174] Step 9:

[0175] The server receives feedback from the user and stores it in a database. The input is the feedback data, and the output is the feedback stored in the database. Specifically, Flask is used to receive the data and perform the storage process.

[0176] Step 10:

[0177] The server then retrains the AI ​​model using the stored feedback. The input is the feedback data in the database, and the output is the retrained AI model. Specifically, it repeats feature extraction and model training as before.

[0178] Step 11:

[0179] The expert inputs his / her insights and analysis reports and sends them to the server. The input is the expert's insights and analysis reports, and the output is the insight data sent to the server. The specific operation is to accept the expert's input via a form.

[0180] Step 12:

[0181] The server stores the insights received from the experts in a database and uses them to retrain the AI ​​model. The input is the expert's insight data, and the output is a retrained AI model that reflects the insights. Specifically, the insight data is stored in a database and used for the re-learning process.

[0182] Step 13:

[0183] The server generates personalized advertisements based on the user's attribute data and displays them on the user's device. The input is the user's attribute data, and the output is personalized advertisements. Specifically, the advertisement content is dynamically changed based on the attribute data and delivered to the user.

[0184] Step 14:

[0185] The server collects user behavior data and evaluates the effectiveness of advertising. The input is user behavior data, and the output is evaluated advertising effectiveness data. Specific operations include analyzing the behavior data and calculating evaluation indicators.

[0186] 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.

[0187] The system for implementing the present invention consists of a multi-step process that combines data collection, data analysis and trend prediction, receiving and processing user feedback, integrating expert insights, presenting prediction results, and an emotion engine that recognizes user emotions.

[0188] Data collection and formatting

[0189] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. It formats this data and converts it into a data frame format. It combines multiple data frames to create a unified dataset and stores this dataset in a database. As an example, it retrieves the latest business news from a news API and converts it into a data frame.

[0190] Feature extraction and AI model training

[0191] The server reads the stored data from the database and extracts the necessary features. An AI model is trained based on the extracted features. This AI model uses algorithms such as random forest and LSTM. Features include market prices, sales data, and seasonal factors.

[0192] Future trend prediction

[0193] Using the trained AI model, the server predicts future trends. The prediction results are displayed in a visually easy-to-understand format (for example, graphs or charts). This allows users to understand the prediction results at a glance. For example, the sales forecast for the next quarter is displayed in chart format.

[0194] Prediction results

[0195] When a user accesses the device, it sends a request to the server to get the latest trend forecast information. The server returns the forecast results to the device, which then displays them on a dashboard in the form of interactive graphs and charts. For example, the UI displays real-time stock price forecasts in infographic format.

[0196] Receiving and processing user feedback

[0197] Users access the system through their own devices and input their predictions and opinions. This feedback is sent to the server and stored in a database. The stored feedback is then used to retrain the AI ​​model. For example, a user's opinion such as "the next sales will increase" is incorporated as retraining data.

[0198] Integrating expert insights

[0199] Experts access the platform and input their own insights and analysis reports, which are then sent to the server and stored in a database.The AI ​​model is then retrained based on the stored insights to improve its prediction accuracy.

[0200] Collecting and processing user sentiment data

[0201] The server collects user emotional data through the emotion engine. For example, when a user enters feedback, the server analyzes emotions from their facial expressions and text to obtain emotional data. This emotional data is then stored in a database along with the user's feedback and opinions.

[0202] Retraining using emotion data

[0203] The server retrains the AI ​​model based on the emotional data obtained from the emotion engine. This retraining process makes it possible to provide prediction results that take the user's emotional state into account, further improving prediction accuracy.

[0204] Specific examples

[0205] For example, the server retrieves business news from a news API, converts it into a data frame, and combines it. Features are extracted and used to train an AI model. Future trends are then predicted and displayed in graph form. When a user accesses the system on their device, this predicted data is displayed on a dashboard. Users input their own opinions and sentiment data, which is sent to the server and used for retraining. Experts also provide insights, improving the overall accuracy of the AI ​​model.

[0206] This is the specific processing flow of this system. By adopting this system, companies will be able to grasp market trends more precisely and in real time, enabling them to make effective decisions.

[0207] The processing flow will be explained below.

[0208] Step 1:

[0209] Data collection and formatting (server)

[0210] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. The retrieved data is converted from JSON format to a data frame format. The server then combines multiple data frames to create a unified dataset and stores this dataset in a database. For example, data retrieved from an economic indicator API is updated daily and stored in a database.

[0211] Step 2:

[0212] Feature extraction (server)

[0213] The server reads the stored data from the database and extracts the necessary features. These features range from sales data, consumer sentiment trends, seasonal factors, product ratings, etc. These features are necessary for the AI ​​model to accurately predict trends.

[0214] Step 3:

[0215] AI model training (server)

[0216] The server trains an AI model based on the extracted features. This AI model uses algorithms such as random forest and LSTM. It uses daily sales data and economic indicators as features and learns the importance of each variable.

[0217] Step 4:

[0218] Future Trend Prediction (Server)

[0219] The server uses the trained AI model to predict future trends. The prediction results are displayed in a visually easy-to-understand format (e.g., graphs and charts), allowing users to understand the prediction results at a glance.

[0220] Step 5:

[0221] Presenting prediction results (device)

[0222] When a user accesses the device, it sends a request to the server to get the latest trend forecast information. The server then sends the forecast results to the device, which then displays them on a dashboard in the form of interactive graphs and charts. For example, the next quarter's sales forecast can be displayed in infographic format.

[0223] Step 6:

[0224] Receiving user feedback (user, device)

[0225] Users access the system through their own terminals and input their predictions and opinions. For example, they can input a prediction such as, "I think sales will increase next quarter." This feedback is sent to the server through a form.

[0226] Step 7:

[0227] Feedback storage and processing (server)

[0228] The server receives the feedback sent by the user and stores it in a database. It then retrains the AI ​​model based on the stored feedback. This retraining process improves the accuracy of the prediction model.

[0229] Step 8:

[0230] Integration of expert insights (server, user)

[0231] Experts access the platform and input their own insights and analytical reports, which the server then stores in a database and uses to retrain the AI ​​model, incorporating the experts' knowledge to further improve its prediction accuracy.

[0232] Step 9:

[0233] Collecting user emotion data (device, user)

[0234] When the user inputs feedback, the device collects the user's emotional data through the emotion engine. For example, when the user inputs feedback, the device analyzes emotions from the user's facial expressions and text to obtain the emotional data.

[0235] Step 10:

[0236] Storage and processing of emotional data (server)

[0237] The server receives the emotion data sent from the device and stores it in a database along with the feedback data. The emotion data is then used to retrain the AI ​​model. This process allows it to provide prediction results that take the user's emotional state into account.

[0238] This is the specific processing flow of this system, which enables companies to make accurate predictions based on multifaceted data, including user sentiment, and supports more effective and faster decision-making.

[0239] Example 2

[0240] 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."

[0241] In today's business environment, accurately predicting market trends is extremely difficult, and companies require real-time analysis and forecasting to make effective decisions. Traditional forecasting systems struggle to take user feedback and sentiment into account, resulting in poor forecast accuracy. Furthermore, the lack of a means to effectively integrate expert insights limits overall forecast accuracy.

[0242] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a data collection means, a means for formatting the collected data and storing it in an information storage device, a means for extracting features based on the stored data and training a generative learning model, a means for predicting future trends using the trained generative learning model, a means for transmitting the prediction results to a terminal device interface and converting them into a displayable format, a means for obtaining feedback from users and retraining the generative learning model, a means for collecting user emotion data using an emotion recognition device, and a means for retraining the generative learning model based on the collected emotion data. This makes it possible to collect and format information from multiple data sources in real time and accurately predict future trends using the generative learning model. Furthermore, taking user feedback and emotions into account improves prediction accuracy, and integrating it with expert insights can further improve prediction accuracy.

[0243] "Data collection means" refers to a means for obtaining data in real time from multiple data sources such as news feeds, social media, and economic indicators.

[0244] An "information storage device" is a database or storage system for saving and managing formatted data.

[0245] "Features" are key data elements extracted from stored data and used to train predictive models.

[0246] A "generative learning model" is an AI model that learns from training data and predicts future trends.

[0247] "Endpoint interface" refers to the interactive display screen or dashboard that a user uses to view prediction results.

[0248] "User feedback" refers to predictions and opinions entered by users through the system.

[0249] "Retraining" is the process of adding new data to an existing training model to improve the model's predictive accuracy.

[0250] An "emotion recognition device" is an engine or system that analyzes and acquires emotional data from user feedback, facial expressions, and text.

[0251] "Emotional data" refers to data that represents a user's emotional state, including emotional categories such as positive, negative, and neutral.

[0252] "Trend forecasting" refers to the process of using trained generative learning models to predict future market movements and trends.

[0253] The present invention is a system for forecasting future market trends by collecting data from multiple data sources in real time, training and relearning a generative learning model based on the collected data, and integrating feedback from servers, terminals, and users to improve forecast accuracy.

[0254] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. It uses common data processing libraries (e.g., Pandas, NumPy) to format this data and convert it into a data frame. It then combines these data frames to create a unified dataset and stores it in a database (e.g., MySQL, PostgreSQL).

[0255] Next, the server reads the stored data from the database and extracts the necessary features. This extraction uses a data analysis library (e.g., Scikit-learn, TensorFlow). Based on the extracted features, a generative learning model (e.g., Random Forest, LSTM) is trained. This training process allows the AI ​​model to learn trends from past data and predict future trends.

[0256] Using the trained generative learning model, the server predicts future trends and formats the prediction results in a visually understandable format such as graphs and charts. This visualization is done using visualization libraries (e.g., Matplotlib, Plotly). The prediction results are sent to the end device interface and displayed on an interactive dashboard when accessed by the user.

[0257] Users access the system through their devices and input their predictions and opinions. This feedback is sent to the server and stored in a database. Based on the stored feedback, the generative learning model is retrained to improve prediction accuracy. This retraining process uses techniques to update the existing model using new data.

[0258] The system also includes an emotion recognizer that collects emotion data from user feedback, facial expressions, and text. Emotion recognition uses natural language processing libraries (e.g., NLTK, spaCy) and face recognition libraries (e.g., OpenCV). The collected emotion data is stored in a database and used to retrain a generative learning model. This allows prediction results to be provided that take the user's emotional state into account.

[0259] As a concrete example, the server retrieves the latest business news from a news API, converts the article contents into a data frame, and combines them. It then extracts features from market prices and performance data to train a generative learning model. It predicts sales for the next quarter and displays the results in a graph format. When accessed by a user, the predicted data is displayed on the device's dashboard, and the user can enter their own feedback and sentiment data. This is sent to the server and used for re-training. Experts also provide insights, improving the accuracy of the generative learning model.

[0260] Prompt Sentence Examples

[0261] "Please use the Random Forest algorithm to predict sales for the next quarter and display the results in a graph."

[0262] "Predict market trends based on sentiment data and present the results to users in infographics."

[0263] "Retrain your generative learning model with user feedback to improve prediction accuracy."

[0264] By adopting this system, companies will be able to collect information from multiple data sources in real time, integrate the opinions of users and experts, and gain a more precise understanding of market trends in real time, enabling them to make more effective decisions.

[0265] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0266] System program processing flow

[0267] Step 1: Data collection

[0268] The server retrieves data in real time from multiple data sources, such as news feeds, social media, and economic indicators. Specifically, it sends API requests to retrieve news articles, social media posts, and economic indicator data. As input, it uses API endpoints and access keys. As output, it returns raw data to the server.

[0269] Step 2: Data Formatting and Storage

[0270] The server formats the retrieved data and converts it into a data frame. Specifically, it extracts information such as the title, content, and date and time of the retrieved news articles and formats them into a data frame using the Pandas or NumPy library. Raw data is used as input, and a data frame is obtained as output. This is then saved in a database. For example, it connects to a database such as MySQL or PostgreSQL and inserts the data.

[0271] Step 3: Feature extraction

[0272] The server reads the stored data from the database and extracts the necessary features. Specifically, features such as market prices, sales data, and seasonal factors are calculated using scripts and extracted using data analysis libraries (e.g., Scikit-learn, TensorFlow). The data read from the database is used as input, and the features are obtained as output.

[0273] Step 4: Training the AI ​​model

[0274] The server trains a generative AI model based on the extracted features. Algorithms used include random forest and LSTM. Specifically, the model is trained using a training dataset and the trained model is saved. The feature dataset is used as input, and the trained model is obtained as output.

[0275] Step 5: Trend forecasting

[0276] The server uses a trained generative AI model to predict future trends. Specifically, it inputs new data into the trained model to predict future market trends and sales. New data is used as input, and predictions are obtained as output. The predictions are then formatted in a visually easy-to-understand format (e.g., graphs or charts).

[0277] Step 6: Formatting and presenting prediction results

[0278] When a user accesses the device, it sends a request to obtain the latest trend forecast information from the server. The server returns the forecast results to the device, which then displays them on a dashboard in the form of interactive graphs and charts. For example, the forecast results can be visualized using libraries such as Plotly or D3.js. The forecast results are used as input, and a visualized dashboard is obtained as output.

[0279] Step 7: Collect and process user feedback

[0280] Users access the system through their own devices and enter their predictions and opinions. A feedback form is provided as input, and users enter their opinions and predictions. This feedback is sent to the server and stored in a database. The saved feedback data is obtained as output.

[0281] Step 8: Integrating expert insights

[0282] Experts access the platform and input their own insights and analytical reports. These insights are sent to the server and stored in the database. The expert insights are used as input and the stored insight data is obtained as output. The generative AI model is then retrained based on the stored insights to improve prediction accuracy.

[0283] Step 9: Collect and process emotion data

[0284] The server uses an emotion recognition device to collect user emotion data. Specifically, when the user enters feedback, it analyzes emotions from their facial expressions and text. It uses libraries such as OpenCV and NLTK. It uses the user's feedback data and facial expression data as input, and obtains analyzed emotion data as output. This is then stored in a database.

[0285] Step 10: Retraining with emotion data

[0286] The server retrains the generative AI model based on the collected emotional data. Specifically, the emotional data is incorporated into the existing learning model as features to improve prediction accuracy. The emotional data and feature data are used as input, and a retrained generative AI model is obtained as output. This provides prediction results that take the user's emotional state into account.

[0287] (Application example 2)

[0288] 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."

[0289] Conventional trend prediction systems have difficulty taking user feedback into account, particularly the influence of user emotions. Furthermore, there is a lack of a means to efficiently integrate real-time user feedback and expert insights, limiting the accuracy of predictions. This makes it difficult to provide product recommendations that meet user needs, especially on online shopping sites.

[0290] 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.

[0291] In this invention, the server includes a data collection means, a means for formatting the collected data and storing it in a database, a means for extracting features from the stored data and training an artificial intelligence model, a means for predicting future trends using the trained artificial intelligence model, a means for sending the prediction results to a user interface and converting them into a displayable format, a means for obtaining feedback from users and retraining the artificial intelligence model, a means for collecting user emotional data and storing it in a database together with the feedback, and a means for retraining the artificial intelligence model based on the emotional data. This enables more accurate trend predictions that take the user's emotional state into account and personalized product recommendations for each user.

[0292] "Data collection means" refers to the ability to obtain information in real time from external data sources.

[0293] "Data formatting means" refers to the function of converting collected raw data into a usable format and storing it in a database.

[0294] "Feature extraction means" refers to the function that extracts important data points from stored data and uses them to train artificial intelligence models.

[0295] An "artificial intelligence model" refers to an algorithm or system that has been trained to make predictions or analyses.

[0296] "Trend Forecasting" refers to the ability to predict future trends using a trained artificial intelligence model.

[0297] "User interface" refers to a function or screen that visually presents prediction results to the user and allows them to operate them interactively.

[0298] "Feedback acquisition means" refers to the function of collecting opinions and advice from users and incorporating them into the system.

[0299] "Re-learning means" refers to the function of re-training the AI ​​model based on collected feedback to improve prediction accuracy.

[0300] "Emotional data" refers to information about a user's emotional state analyzed from their text and images.

[0301] "Emotional data collection means" refers to the function of analyzing the user's emotional state and storing it in a database.

[0302] The system for implementing the present invention consists of the following steps: data collection, data analysis and trend prediction, receiving and processing user feedback, collecting and integrating emotion data, and re-training. Each step is described in detail below.

[0303] Data collection

[0304] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. For example, news APIs, social media APIs, and economic indicator APIs are used. These raw data are converted into a data frame format and combined to create a unified dataset. This dataset is then stored in a database system, such as MySQL or PostgreSQL.

[0305] Feature extraction and AI model training

[0306] The server reads the stored data from the database and extracts the necessary features. Technically, this can be done using Python data analysis libraries such as pandas or numpy. Based on the extracted features, an artificial intelligence model is trained using a machine learning library such as scikit-learn. This AI model can use, for example, random forest or LSTM.

[0307] Future trend prediction

[0308] Using trained artificial intelligence models, the server predicts future trends. The results are then formatted into easy-to-understand visual displays (e.g., graphs and charts). For example, the next quarter's sales forecast is displayed in chart format.

[0309] Prediction results

[0310] When a user accesses the device, it sends a request to the server to get the latest trend forecast information. The server returns the forecast results to the device, which then displays them on a dashboard in the form of interactive graphs and charts. For example, the UI displays real-time stock price forecasts in infographic format.

[0311] Receiving and processing user feedback

[0312] Users access the system through their devices and input their predictions and opinions. This feedback is entered, for example, using an HTML form and sent to the server. The server receives it and stores it in a database. The stored feedback is then used to retrain the AI ​​model.

[0313] Collecting and processing emotional data

[0314] The server collects user emotion data through an emotion recognition engine. For example, libraries such as TextBlob and facial_emotion_recognition are used to analyze the emotion of facial expressions and text when users enter feedback. The resulting emotion data is then stored in a database along with the feedback.

[0315] Retraining using emotion data

[0316] The server retrains the AI ​​model based on the emotion data obtained from the emotion engine. This retraining process allows it to provide prediction results that take the user's emotional state into account, further improving prediction accuracy.

[0317] As a concrete example, the server retrieves business news from a news API, converts it into a data frame, and combines it. Features are extracted and an AI model is trained. Next, future trends are predicted and displayed in graph form. When a user accesses the system on their device, this predicted data is displayed on a dashboard. The user enters emotional data along with their own opinion, which is sent to the server and used for re-learning. An example of a prompt used here is, "Do you think sales will increase in the next quarter? Please explain why. Please also upload an image of your feedback."

[0318] The above is a specific embodiment of the present system. By adopting this system, companies can grasp market trends more precisely and in real time, enabling them to make effective decisions.

[0319] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0320] Step 1:

[0321] The server sends API requests from multiple data sources, such as news feeds, social media, and economic indicators, and retrieves data in real time.

[0322] Input: API request

[0323] Data processing: Convert the response from each API into a data frame format

[0324] Output: Data in data frame format

[0325] Step 2:

[0326] The server formats the acquired data into multiple data frames and combines them to create a unified dataset.

[0327] Input: Dataframe of news, social media, and economic indicator data

[0328] Data processing: Combining data frames

[0329] Output: Unified dataset

[0330] Step 3:

[0331] The server stores the unified data set in a database.

[0332] Input: Unified dataset

[0333] Action: Insert data into database

[0334] Output: Saved

[0335] Step 4:

[0336] The server reads the stored data from the database and extracts the necessary features.

[0337] Input: Data read from database

[0338] Data processing: feature extraction process

[0339] Output: Extracted features

[0340] Step 5:

[0341] The server trains an artificial intelligence model based on the extracted features.

[0342] Input: extracted features

[0343] Data processing: training artificial intelligence models

[0344] Output: A trained artificial intelligence model

[0345] Step 6:

[0346] The server uses trained artificial intelligence models to predict future trends.

[0347] Input: A trained AI model and the data to be predicted

[0348] Data calculation: Future trend prediction process

[0349] Output: Prediction results

[0350] Step 7:

[0351] The server formats the prediction results into interactive graphs and charts and sends them to the user interface.

[0352] Input: Prediction result

[0353] Data processing: Conversion into infographics format

[0354] Output: Interactive graphs and charts

[0355] Step 8:

[0356] The device displays the prediction results in the user interface and is ready to receive feedback from the user.

[0357] Input: Interactive graphs and charts

[0358] Behavior: Show in user interface

[0359] Output: Displayed prediction results

[0360] Step 9:

[0361] Users enter their predictions and opinions through a form and also upload image feedback.

[0362] Input: User opinions and feedback images

[0363] Action: Enter and send feedback

[0364] Output: Feedback data

[0365] Step 10:

[0366] The server performs text analysis on the received feedback to obtain emotion data.

[0367] Input: User feedback

[0368] Data Computing: Text Analysis and Emotion Recognition

[0369] Output: Emotion data

[0370] Step 11:

[0371] The server stores the acquired emotion data and feedback in a database.

[0372] Input: Emotion data and feedback

[0373] Action: Insert data into database

[0374] Output: Saved

[0375] Step 12:

[0376] The server retrains the artificial intelligence model based on the stored feedback and emotion data.

[0377] Input: Feedback and emotion data

[0378] Data Computing: Retraining Artificial Intelligence Models

[0379] Output: Updated artificial intelligence model

[0380] The above is the specific operation and data flow of each processing step in this system.

[0381] 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.

[0382] 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.

[0383] 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.

[0384] [Second embodiment]

[0385] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0386] 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.

[0387] 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).

[0388] 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.

[0389] 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.

[0390] 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).

[0391] 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.

[0392] 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.

[0393] 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.

[0394] 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.

[0395] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0396] 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."

[0397] A system for implementing the present invention combines a multi-step process that includes data collection, data analysis and trend prediction, receiving and processing user feedback, integrating expert insights, and presenting the predictive results.

[0398] Data collection

[0399] The server first collects data in real time. To do this, it obtains data from multiple data sources, such as news feeds, social media, and economic indicators, via APIs. The obtained data is then formatted and converted into a data frame. Multiple data frames are then combined to create a unified dataset. This unified dataset is then stored in a database. This process requires efficient data collection and storage.

[0400] Data analysis and trend forecasting

[0401] The server then reads the stored data from the database and extracts the necessary features. It then trains the data using an AI model, which uses algorithms such as random forest and LSTM. Once trained, the model is used to predict future trends. The prediction results are then formatted so that they are displayed to the user in a visually understandable format and sent to the user's device.

[0402] Receiving and processing feedback

[0403] Users access the system through their devices and input their predictions and opinions. For example, they can enter a prediction such as, "I think sales will increase next quarter." These predictions and opinions are sent to the server, which stores the received feedback in a database and further uses this feedback to update the training data and retrain the AI ​​model.

[0404] Integrating expert insights

[0405] Experts access the platform and input their own insights and analysis reports. These insights are sent to the server and stored in a database. The AI ​​model is then retrained based on the stored insights to improve its prediction accuracy. This process allows expert knowledge to be incorporated into the model.

[0406] Prediction results

[0407] When a user accesses the terminal, it sends a request to the server to obtain the latest trend forecast information. The server returns the latest forecast results, and the terminal displays these results on the dashboard in the form of interactive graphs and charts, allowing users to grasp the latest market trends at a glance.

[0408] As described above, the present invention provides a system that consistently performs processes from data collection to prediction and feedback integration, thereby enabling companies to make more effective and faster decisions and supporting them in responding quickly to market changes.

[0409] The processing flow will be explained below.

[0410] Step 1:

[0411] Data collection and formatting (server)

[0412] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. The data is then formatted and converted into a data frame. These data frames are then combined to create a unified dataset. Finally, this unified dataset is stored in a database.

[0413] Step 2:

[0414] Feature extraction (server)

[0415] The server reads the stored data from the database and extracts the necessary features. Features are the data attributes and variables required to train a predictive model. These are important elements for correctly forecasting market trends.

[0416] Step 3:

[0417] AI model training (server)

[0418] The server trains an AI model based on the extracted features. Models used here include random forests and LSTM. The server uses these algorithms to learn the training data. Cross-validation and other methods are performed to improve the model's performance.

[0419] Step 4:

[0420] Future Trend Prediction (Server)

[0421] The server uses the trained AI model to predict future trends, and the prediction results are displayed in a visually easy-to-understand format (e.g., graphs and charts), allowing users to understand the prediction results at a glance.

[0422] Step 5:

[0423] Presenting prediction results (device)

[0424] When a user accesses the terminal, it sends a request to the server to obtain the latest trend forecast information. The server returns the forecast results to the terminal, which then displays them on the dashboard in the form of interactive graphs and charts, allowing users to understand market trends in real time.

[0425] Step 6:

[0426] Receiving user feedback (device, user)

[0427] Users input their predictions and opinions through their own devices. For example, they input predictions such as, "I think sales will increase in the next quarter." This feedback is sent to the server through a form.

[0428] Step 7:

[0429] Feedback storage and processing (server)

[0430] The server receives the feedback sent by the user and stores it in a database, after which the AI ​​model is retrained based on the stored feedback, a process that allows the accuracy of the prediction model to be improved.

[0431] Step 8:

[0432] Integration of expert insights (server, user)

[0433] Experts access the platform and input their own insights and analysis reports. The server stores these insights in a database and uses them to retrain the AI ​​model. By incorporating the experts' knowledge into the model, the accuracy of predictions can be further improved.

[0434] The above is the specific processing flow of this system.

[0435] Example 1

[0436] 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."

[0437] Traditional data analysis systems struggled to consistently collect data, forecast trends, and integrate feedback. This made it difficult for companies to obtain the latest market trend information in a timely manner to make quick decisions. Furthermore, there was a lack of a way to effectively integrate feedback and insights from users and experts to improve forecast accuracy.

[0438] 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.

[0439] In this invention, the server includes a data collection means, a means for formatting the collected data and storing it in a database, a means for extracting features from the stored data and training a machine learning model, a means for predicting future trends using the trained machine learning model, a means for sending the prediction results to a user interface and converting them into a displayable format, a means for obtaining feedback from users and retraining the machine learning model, a means for receiving expert insights through the platform and storing them in a database, and a means for retraining the machine learning model based on the stored expert insights to improve prediction accuracy. This enables companies to collect data in real time, make highly accurate trend predictions, and build a feedback loop that reflects the opinions of users and experts.

[0440] "Data collection means" refers to the means for obtaining data in real time from various data sources such as news feeds, social media, and economic indicators.

[0441] "Format" is the process of converting collected data into a usable format and carrying out the data cleaning and formatting required for storing it in a database.

[0442] "Means for storing in a database" refers to the means for storing formatted data in a database so that it can be efficiently searched and retrieved.

[0443] "Features" are attributes and values ​​of data extracted and transformed to be fed into a machine learning model, and contain information that is important for analysis and prediction.

[0444] A "machine learning model" is an algorithm or model that is trained using data to make predictions or classifications based on future data.

[0445] "Training means" refers to the means for training a machine learning model using collected data to improve the accuracy of the model.

[0446] A "means for predicting future trends" is a means for predicting trends in future data or events using a trained machine learning model.

[0447] A "user interface" is an interface through which a user accesses a system to input information or obtain information from the system.

[0448] The "means for converting into a displayable format" refers to a means for converting the prediction results into a visually easy-to-understand format (for example, a graph or chart) and displaying it on a user interface.

[0449] "Feedback" refers to predictions, opinions, and insights provided by users and experts.

[0450] "Means for obtaining feedback" refers to the means for collecting and storing feedback from users and experts.

[0451] "Retraining" refers to the process of retraining a machine learning model using feedback and new data to improve the model's accuracy.

[0452] "Platform" means a designated environment or system for experts to input and integrate their insights and analytical reports into the system.

[0453] "Insights" refers to detailed analysis and opinions provided by experts, and are important information for improving forecast accuracy.

[0454] The system of the present invention combines a multi-step process, including data collection, data analysis and trend prediction, receiving and processing user feedback, integrating expert insights, and presenting the prediction results. As a specific implementation method of the present invention, the following steps are described in detail.

[0455] Data collection

[0456] The server collects data in real time from multiple data sources, such as news feeds, social media, and economic indicators, via APIs. For example, the server uses Twitter's API to collect tweets based on specific keywords. It converts this data into a data frame using the Pandas library. The server then combines multiple data frames to create a unified dataset, which is then stored in an SQL database.

[0457] Data analysis and trend forecasting

[0458] The server reads the stored data from the database and extracts features. For example, the server extracts the number of tweets in the past 24 hours and the percentage of tweets with positive context. The server then trains a machine learning model such as random forest or LSTM using Scikit-learn or TensorFlow. The trained model is used to predict future trends. The prediction results are formatted into a visually easy-to-understand format (e.g., graphs or charts) using the Plotly library and sent to the user's device.

[0459] Receiving and processing feedback

[0460] Users use their own devices to input their predictions and opinions into the system. For example, a user can input a prediction such as, "I think sales will increase next quarter." The device then sends the input feedback to the server. The server stores the received feedback in a database and updates the training data based on the stored feedback. The server then retrains the machine learning model using the updated data.

[0461] Integrating expert insights

[0462] Experts access the platform and input their own insights and analysis reports. The device then sends these insights to the server, which stores them in a database and retrains the machine learning model based on the stored insights. This incorporates expert knowledge into the model, improving its prediction accuracy.

[0463] Prediction results

[0464] The user's device sends a request to the server to get the latest trend forecast information, and the server returns the latest forecast results, which the device displays on a dashboard in the form of interactive graphs and charts.

[0465] Specific examples

[0466] For example, a user can enter the following prompt into the system:

[0467] "Collect the latest tweet data and make supply and demand forecasts."

[0468] "Create a sales forecast for the next quarter and display it in a chart."

[0469] "Update the AI ​​model based on user feedback."

[0470] In response to these prompts, the server collects, analyzes, and processes appropriate data. Furthermore, by incorporating expert insights into the model, forecast accuracy can be improved. In this way, the system of the present invention helps companies respond quickly to market changes and make effective decisions.

[0471] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0472] Step 1: Data collection

[0473] The server uses APIs to collect data in real time from various data sources, such as news feeds, social media, and economic indicators. For example, the Twitter API is used to collect tweets based on specific keywords. The server periodically executes scheduled tasks to obtain the data. The collected data is converted into a data frame using the Pandas library. The input is raw data from each data source, and the output is a formatted data frame.

[0474] Step 2: Format and save the data

[0475] The server reformats the collected data using the Pandas library. This reformatting process includes removing duplicate data, handling missing values, and normalizing text. The reformatted data is then combined to create a unified dataset. This dataset is then stored in a SQL database. The input is an unformed data frame, and the output is the reformatted and combined dataset.

[0476] Step 3: Feature extraction

[0477] The server reads the stored data from the database and extracts features, such as the number of tweets per hour or the percentage of tweets with positive context. The Scikit-learn library is used to encode these features. The input is the dataset, and the output is a list of numerically encoded features.

[0478] Step 4: Train the model

[0479] The server trains the extracted features using machine learning models such as random forests and LSTM. For example, it uses past tweet data to predict future trends. Libraries such as TensorFlow and Scikit-learn are used for this training. The accuracy of the trained model is verified and saved. The input is a list of features, and the output is a trained machine learning model.

[0480] Step 5: Predicting trends

[0481] The server uses the trained model to predict future trends, for example, forecasting sales for the next quarter. The prediction results are visualized using the Plotly library and output in a format that is easy for users to understand. The input is the machine learning model and features, and the output is the visualized prediction results.

[0482] Step 6: Receiving feedback

[0483] Users input their predictions and opinions into the system through their terminals. For example, they input a prediction such as, "I think sales will increase next quarter." The terminals collect this feedback and send it to the server. The input is the user's feedback, and the output is the feedback data sent to the server.

[0484] Step 7: Process feedback and retrain

[0485] The server stores the received feedback in a database and updates the training data based on the stored feedback. The AI ​​model is retrained using the updated data, which improves the model's accuracy. The input is the feedback data, and the output is an updated machine learning model.

[0486] Step 8: Integrating expert insights

[0487] Experts access the platform and input their insights and analysis reports. The device sends them to the server, which stores the received insights in a database. The stored insights are used to retrain the machine learning model, further improving its prediction accuracy. The input is the expert's insights, and the output is the retrained machine learning model.

[0488] Step 9: Presenting the prediction results

[0489] The user's device requests the latest forecast results from the server. The server returns the latest forecast results, and the device displays these results on a dashboard as interactive graphs and charts, allowing the user to understand the latest market trends at a glance. The input is a request for forecast results, and the output is the forecast results in a displayable format.

[0490] (Application example 1)

[0491] 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."

[0492] Conventional ad delivery systems lacked real-time market trend analysis, making effective targeting and personalization difficult. They also struggled to quickly incorporate user and expert feedback, delaying the optimization of ad campaigns. Furthermore, limited means of collecting and analyzing user behavior data made it difficult to accurately evaluate advertising effectiveness.

[0493] 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.

[0494] In this invention, the server includes a data collection means, a means for formatting the collected data and storing it in a database, a means for extracting features from the stored data and training an AI model, a means for predicting future trends using the trained AI model, a means for sending the prediction results to a user interface and converting them into a displayable format, a means for receiving feedback from users and retraining the AI ​​model, a means for receiving insights and analysis reports from experts and storing them in a database and retraining the AI ​​model, and a means for generating personalized advertisements based on the collected and analyzed data and displaying them on user terminals. This enables effective targeting based on real-time market trend analysis, optimization of advertising campaigns that quickly reflect user and expert feedback, and accurate evaluation of advertising effectiveness by collecting and analyzing user behavior data.

[0495] "Data collection means" refers to a means for obtaining data in real time from multiple data sources such as news feeds, social media, and economic indicators.

[0496] "Database" means a centralized data repository for storing uniform data sets for later use in analysis and prediction.

[0497] A "feature extraction method" is a method for identifying information necessary for trend prediction from collected data and converting it into a format that can be used to train an AI model.

[0498] "AI model training method" refers to a method of training an AI model (such as random forest or LSTM) using extracted features to enable it to predict future trends.

[0499] A "trend forecasting tool" is a tool for forecasting future trends using a trained AI model.

[0500] The "user interface transmission means" is a means for converting the prediction results into a format that is easy for the user to understand and transmitting it to the user's terminal.

[0501] "Feedback acquisition means" refers to the means of collecting predictions and opinions from users and reflecting them in the system.

[0502] "AI model retraining method" refers to a method of retraining an AI model based on feedback from users and experts to improve its accuracy.

[0503] The "means for receiving expert insights" is a means for receiving insights and analysis reports from experts, storing them in a database, and then retraining the AI ​​model.

[0504] "Personalized advertisement generation means" refers to a means for generating advertisements optimized for individual users based on collected and analyzed data and displaying them on the user's terminal.

[0505] The system that realizes this application example comprises the smartphone app "AdTrend." The server collects data in real time from multiple data sources, such as news feeds, social media, and economic indicators. The collected data is formatted and converted into a data frame, and these data frames are then combined to create a unified dataset. This process uses Python's Pandas framework.

[0506] Next, the server reads the stored data from the database and extracts the necessary features. Based on the extracted features, it trains an AI model using algorithms such as random forest and LSTM. This training generates an AI model that can predict future trends. The model is trained using the Scikit-learn library and generates prediction results.

[0507] The server then formats these forecasts in JSON format and sends them to the user's smartphone, where they are displayed as interactive charts and graphs. This dashboard helps users visually understand trends.

[0508] Users can enter their predictions and opinions into the system through a form. This feedback is sent to the server and stored in a database. The stored feedback is used as new training data to retrain the AI ​​model. The Flask framework is used to receive and process user feedback.

[0509] In addition, insights and analysis reports from experts are also sent to the server and stored in the database, allowing the experts' knowledge to be utilized to improve the prediction accuracy of the AI ​​model. In this process, the experts' insights are incorporated into the retraining of the AI ​​model.

[0510] Regarding the generation of personalized ads, the collected data is used to generate ads optimized for individual user attributes. These ads are delivered and displayed on the user's device. User behavior data is collected and analyzed to evaluate the effectiveness of the ads. Ad campaigns are optimized using Python and related libraries.

[0511] Examples of specific prompts include:

[0512] Trend Forecast:

[0513] Analyze the characteristics of the most popular advertising campaigns in the next quarter and propose effective advertising strategies.

[0514] Feedback received:

[0515] We collect user opinions about the ads currently being served and use that data to personalize ads.

[0516] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0517] Step 1:

[0518] The server collects data in real time from multiple data sources, such as news feeds, social media, and economic indicators. It uses an API to retrieve the data and converts it into a Pandas data frame. The input is raw data from each data source, and the output is organized data in a data frame format.

[0519] Step 2:

[0520] The server combines the multiple formatted data frames to create a unified dataset and stores it in the database. The input is multiple data frames, and the output is the combined unified dataset. Specifically, the server combines the data frames using functions such as the merge function in Pandas and stores it in the database.

[0521] Step 3:

[0522] The server reads the stored data from the database and extracts the necessary features. The input is a unified dataset, and the output is the features required for training the AI ​​model. Specifically, it uses Pandas and Scikit-learn to extract the features and prepare them as training data.

[0523] Step 4:

[0524] The server uses the extracted features to train an AI model. The input is the features, and the output is a trained AI model (e.g., random forest or LSTM). Specifically, the server uses the Scikit-learn library to train the model and check its prediction accuracy.

[0525] Step 5:

[0526] The server uses a trained AI model to predict future trends. The input is newly collected data, and the output is the future trend prediction result. Specifically, new data is input into the model and a prediction result is generated.

[0527] Step 6:

[0528] The server formats the prediction results in JSON format and sends them to the user's device. The input is the prediction result data, and the output is the JSON-formatted prediction result sent to the user's device. Specifically, Flask is used to communicate data between the server and the device.

[0529] Step 7:

[0530] The terminal displays the received prediction results as interactive charts and graphs. The input is the prediction result data in JSON format, and the output is the prediction result displayed graphically. Specifically, a dashboard is created using a JavaScript library (e.g., D3.js) to provide visual feedback to the user.

[0531] Step 8:

[0532] Users enter their predictions and opinions into the system through a form. The input is the user's feedback, and the output is the feedback data sent to the server. Specifically, the system uses a combination of HTML forms and JavaScript to send user input to the server.

[0533] Step 9:

[0534] The server receives feedback from the user and stores it in a database. The input is the feedback data, and the output is the feedback stored in the database. Specifically, Flask is used to receive the data and perform the storage process.

[0535] Step 10:

[0536] The server then retrains the AI ​​model using the stored feedback. The input is the feedback data in the database, and the output is the retrained AI model. Specifically, it repeats feature extraction and model training as before.

[0537] Step 11:

[0538] The expert inputs his / her insights and analysis reports and sends them to the server. The input is the expert's insights and analysis reports, and the output is the insight data sent to the server. The specific operation is to accept the expert's input via a form.

[0539] Step 12:

[0540] The server stores the insights received from the experts in a database and uses them to retrain the AI ​​model. The input is the expert's insight data, and the output is a retrained AI model that reflects the insights. Specifically, the insight data is stored in a database and used for the re-learning process.

[0541] Step 13:

[0542] The server generates personalized advertisements based on the user's attribute data and displays them on the user's device. The input is the user's attribute data, and the output is personalized advertisements. Specifically, the advertisement content is dynamically changed based on the attribute data and delivered to the user.

[0543] Step 14:

[0544] The server collects user behavior data and evaluates the effectiveness of advertising. The input is user behavior data, and the output is evaluated advertising effectiveness data. Specific operations include analyzing the behavior data and calculating evaluation indicators.

[0545] 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.

[0546] The system for implementing the present invention consists of a multi-step process that combines data collection, data analysis and trend prediction, receiving and processing user feedback, integrating expert insights, presenting prediction results, and an emotion engine that recognizes user emotions.

[0547] Data collection and formatting

[0548] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. It formats this data and converts it into a data frame format. It combines multiple data frames to create a unified dataset and stores this dataset in a database. As an example, it retrieves the latest business news from a news API and converts it into a data frame.

[0549] Feature extraction and AI model training

[0550] The server reads the stored data from the database and extracts the necessary features. An AI model is trained based on the extracted features. This AI model uses algorithms such as random forest and LSTM. Features include market prices, sales data, and seasonal factors.

[0551] Future trend prediction

[0552] Using the trained AI model, the server predicts future trends. The prediction results are displayed in a visually easy-to-understand format (for example, graphs or charts). This allows users to understand the prediction results at a glance. For example, the sales forecast for the next quarter is displayed in chart format.

[0553] Prediction results

[0554] When a user accesses the device, it sends a request to the server to get the latest trend forecast information. The server returns the forecast results to the device, which then displays them on a dashboard in the form of interactive graphs and charts. For example, the UI displays real-time stock price forecasts in infographic format.

[0555] Receiving and processing user feedback

[0556] Users access the system through their own devices and input their predictions and opinions. This feedback is sent to the server and stored in a database. The stored feedback is then used to retrain the AI ​​model. For example, a user's opinion such as "the next sales will increase" is incorporated as retraining data.

[0557] Integrating expert insights

[0558] Experts access the platform and input their own insights and analysis reports, which are then sent to the server and stored in a database.The AI ​​model is then retrained based on the stored insights to improve its prediction accuracy.

[0559] Collecting and processing user sentiment data

[0560] The server collects user emotional data through the emotion engine. For example, when a user enters feedback, the server analyzes emotions from their facial expressions and text to obtain emotional data. This emotional data is then stored in a database along with the user's feedback and opinions.

[0561] Retraining using emotion data

[0562] The server retrains the AI ​​model based on the emotional data obtained from the emotion engine. This retraining process makes it possible to provide prediction results that take the user's emotional state into account, further improving prediction accuracy.

[0563] Specific examples

[0564] For example, the server retrieves business news from a news API, converts it into a data frame, and combines it. Features are extracted and used to train an AI model. Future trends are then predicted and displayed in graph form. When a user accesses the system on their device, this predicted data is displayed on a dashboard. Users input their own opinions and sentiment data, which is sent to the server and used for retraining. Experts also provide insights, improving the overall accuracy of the AI ​​model.

[0565] This is the specific processing flow of this system. By adopting this system, companies will be able to grasp market trends more precisely and in real time, enabling them to make effective decisions.

[0566] The processing flow will be explained below.

[0567] Step 1:

[0568] Data collection and formatting (server)

[0569] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. The retrieved data is converted from JSON format to a data frame format. The server then combines multiple data frames to create a unified dataset and stores this dataset in a database. For example, data retrieved from an economic indicator API is updated daily and stored in a database.

[0570] Step 2:

[0571] Feature extraction (server)

[0572] The server reads the stored data from the database and extracts the necessary features. These features range from sales data, consumer sentiment trends, seasonal factors, product ratings, etc. These features are necessary for the AI ​​model to accurately predict trends.

[0573] Step 3:

[0574] AI model training (server)

[0575] The server trains an AI model based on the extracted features. This AI model uses algorithms such as random forest and LSTM. It uses daily sales data and economic indicators as features and learns the importance of each variable.

[0576] Step 4:

[0577] Future Trend Prediction (Server)

[0578] The server uses the trained AI model to predict future trends. The prediction results are displayed in a visually easy-to-understand format (e.g., graphs and charts), allowing users to understand the prediction results at a glance.

[0579] Step 5:

[0580] Presenting prediction results (device)

[0581] When a user accesses the device, it sends a request to the server to get the latest trend forecast information. The server then sends the forecast results to the device, which then displays them on a dashboard in the form of interactive graphs and charts. For example, the next quarter's sales forecast can be displayed in infographic format.

[0582] Step 6:

[0583] Receiving user feedback (user, device)

[0584] Users access the system through their own terminals and input their predictions and opinions. For example, they can input a prediction such as, "I think sales will increase next quarter." This feedback is sent to the server through a form.

[0585] Step 7:

[0586] Feedback storage and processing (server)

[0587] The server receives the feedback sent by the user and stores it in a database. It then retrains the AI ​​model based on the stored feedback. This retraining process improves the accuracy of the prediction model.

[0588] Step 8:

[0589] Integration of expert insights (server, user)

[0590] Experts access the platform and input their own insights and analytical reports, which the server then stores in a database and uses to retrain the AI ​​model, incorporating the experts' knowledge to further improve its prediction accuracy.

[0591] Step 9:

[0592] Collecting user emotion data (device, user)

[0593] When the user inputs feedback, the device collects the user's emotional data through the emotion engine. For example, when the user inputs feedback, the device analyzes emotions from the user's facial expressions and text to obtain the emotional data.

[0594] Step 10:

[0595] Storage and processing of emotional data (server)

[0596] The server receives the emotion data sent from the device and stores it in a database along with the feedback data. The emotion data is then used to retrain the AI ​​model. This process allows it to provide prediction results that take the user's emotional state into account.

[0597] This is the specific processing flow of this system, which enables companies to make accurate predictions based on multifaceted data, including user sentiment, and supports more effective and faster decision-making.

[0598] Example 2

[0599] 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."

[0600] In today's business environment, accurately predicting market trends is extremely difficult, and companies require real-time analysis and forecasting to make effective decisions. Traditional forecasting systems struggle to take user feedback and sentiment into account, resulting in poor forecast accuracy. Furthermore, the lack of a means to effectively integrate expert insights limits overall forecast accuracy.

[0601] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a data collection means, a means for formatting the collected data and storing it in an information storage device, a means for extracting features based on the stored data and training a generative learning model, a means for predicting future trends using the trained generative learning model, a means for transmitting the prediction results to a terminal device interface and converting them into a displayable format, a means for obtaining feedback from users and retraining the generative learning model, a means for collecting user emotion data using an emotion recognition device, and a means for retraining the generative learning model based on the collected emotion data. This makes it possible to collect and format information from multiple data sources in real time and accurately predict future trends using the generative learning model. Furthermore, taking user feedback and emotions into account improves prediction accuracy, and integrating it with expert insights can further improve prediction accuracy.

[0602] "Data collection means" refers to a means for obtaining data in real time from multiple data sources such as news feeds, social media, and economic indicators.

[0603] An "information storage device" is a database or storage system for saving and managing formatted data.

[0604] "Features" are key data elements extracted from stored data and used to train predictive models.

[0605] A "generative learning model" is an AI model that learns from training data and predicts future trends.

[0606] "Endpoint interface" refers to the interactive display screen or dashboard that a user uses to view prediction results.

[0607] "User feedback" refers to predictions and opinions entered by users through the system.

[0608] "Retraining" is the process of adding new data to an existing training model to improve the model's predictive accuracy.

[0609] An "emotion recognition device" is an engine or system that analyzes and acquires emotional data from user feedback, facial expressions, and text.

[0610] "Emotional data" refers to data that represents a user's emotional state, including emotional categories such as positive, negative, and neutral.

[0611] "Trend forecasting" refers to the process of using trained generative learning models to predict future market movements and trends.

[0612] The present invention is a system for forecasting future market trends by collecting data from multiple data sources in real time, training and relearning a generative learning model based on the collected data, and integrating feedback from servers, terminals, and users to improve forecast accuracy.

[0613] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. It uses common data processing libraries (e.g., Pandas, NumPy) to format this data and convert it into a data frame. It then combines these data frames to create a unified dataset and stores it in a database (e.g., MySQL, PostgreSQL).

[0614] Next, the server reads the stored data from the database and extracts the necessary features. This extraction uses a data analysis library (e.g., Scikit-learn, TensorFlow). Based on the extracted features, a generative learning model (e.g., Random Forest, LSTM) is trained. This training process allows the AI ​​model to learn trends from past data and predict future trends.

[0615] Using the trained generative learning model, the server predicts future trends and formats the prediction results in a visually understandable format such as graphs and charts. This visualization is done using visualization libraries (e.g., Matplotlib, Plotly). The prediction results are sent to the end device interface and displayed on an interactive dashboard when accessed by the user.

[0616] Users access the system through their devices and input their predictions and opinions. This feedback is sent to the server and stored in a database. Based on the stored feedback, the generative learning model is retrained to improve prediction accuracy. This retraining process uses techniques to update the existing model using new data.

[0617] The system also includes an emotion recognizer that collects emotion data from user feedback, facial expressions, and text. Emotion recognition uses natural language processing libraries (e.g., NLTK, spaCy) and face recognition libraries (e.g., OpenCV). The collected emotion data is stored in a database and used to retrain a generative learning model. This allows prediction results to be provided that take the user's emotional state into account.

[0618] As a concrete example, the server retrieves the latest business news from a news API, converts the article contents into a data frame, and combines them. It then extracts features from market prices and performance data to train a generative learning model. It predicts sales for the next quarter and displays the results in a graph format. When accessed by a user, the predicted data is displayed on the device's dashboard, and the user can enter their own feedback and sentiment data. This is sent to the server and used for re-training. Experts also provide insights, improving the accuracy of the generative learning model.

[0619] Prompt Sentence Examples

[0620] "Please use the Random Forest algorithm to predict sales for the next quarter and display the results in a graph."

[0621] "Predict market trends based on sentiment data and present the results to users in infographics."

[0622] "Retrain your generative learning model with user feedback to improve prediction accuracy."

[0623] By adopting this system, companies will be able to collect information from multiple data sources in real time, integrate the opinions of users and experts, and gain a more precise understanding of market trends in real time, enabling them to make more effective decisions.

[0624] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0625] System program processing flow

[0626] Step 1: Data collection

[0627] The server retrieves data in real time from multiple data sources, such as news feeds, social media, and economic indicators. Specifically, it sends API requests to retrieve news articles, social media posts, and economic indicator data. As input, it uses API endpoints and access keys. As output, it returns raw data to the server.

[0628] Step 2: Data Formatting and Storage

[0629] The server formats the retrieved data and converts it into a data frame. Specifically, it extracts information such as the title, content, and date and time of the retrieved news articles and formats them into a data frame using the Pandas or NumPy library. Raw data is used as input, and a data frame is obtained as output. This is then saved in a database. For example, it connects to a database such as MySQL or PostgreSQL and inserts the data.

[0630] Step 3: Feature extraction

[0631] The server reads the stored data from the database and extracts the necessary features. Specifically, features such as market prices, sales data, and seasonal factors are calculated using scripts and extracted using data analysis libraries (e.g., Scikit-learn, TensorFlow). The data read from the database is used as input, and the features are obtained as output.

[0632] Step 4: Training the AI ​​model

[0633] The server trains a generative AI model based on the extracted features. Algorithms used include random forest and LSTM. Specifically, the model is trained using a training dataset and the trained model is saved. The feature dataset is used as input, and the trained model is obtained as output.

[0634] Step 5: Trend forecasting

[0635] The server uses a trained generative AI model to predict future trends. Specifically, it inputs new data into the trained model to predict future market trends and sales. New data is used as input, and predictions are obtained as output. The predictions are then formatted in a visually easy-to-understand format (e.g., graphs or charts).

[0636] Step 6: Formatting and presenting prediction results

[0637] When a user accesses the device, it sends a request to obtain the latest trend forecast information from the server. The server returns the forecast results to the device, which then displays them on a dashboard in the form of interactive graphs and charts. For example, the forecast results can be visualized using libraries such as Plotly or D3.js. The forecast results are used as input, and a visualized dashboard is obtained as output.

[0638] Step 7: Collect and process user feedback

[0639] Users access the system through their own devices and enter their predictions and opinions. A feedback form is provided as input, and users enter their opinions and predictions. This feedback is sent to the server and stored in a database. The saved feedback data is obtained as output.

[0640] Step 8: Integrating expert insights

[0641] Experts access the platform and input their own insights and analytical reports. These insights are sent to the server and stored in the database. The expert insights are used as input and the stored insight data is obtained as output. The generative AI model is then retrained based on the stored insights to improve prediction accuracy.

[0642] Step 9: Collect and process emotion data

[0643] The server uses an emotion recognition device to collect user emotion data. Specifically, when the user enters feedback, it analyzes emotions from their facial expressions and text. It uses libraries such as OpenCV and NLTK. It uses the user's feedback data and facial expression data as input, and obtains analyzed emotion data as output. This is then stored in a database.

[0644] Step 10: Retraining with emotion data

[0645] The server retrains the generative AI model based on the collected emotional data. Specifically, the emotional data is incorporated into the existing learning model as features to improve prediction accuracy. The emotional data and feature data are used as input, and a retrained generative AI model is obtained as output. This provides prediction results that take the user's emotional state into account.

[0646] (Application example 2)

[0647] 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."

[0648] Conventional trend prediction systems have difficulty taking user feedback into account, particularly the influence of user emotions. Furthermore, there is a lack of a means to efficiently integrate real-time user feedback and expert insights, limiting the accuracy of predictions. This makes it difficult to provide product recommendations that meet user needs, especially on online shopping sites.

[0649] 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.

[0650] In this invention, the server includes a data collection means, a means for formatting the collected data and storing it in a database, a means for extracting features from the stored data and training an artificial intelligence model, a means for predicting future trends using the trained artificial intelligence model, a means for sending the prediction results to a user interface and converting them into a displayable format, a means for obtaining feedback from users and retraining the artificial intelligence model, a means for collecting user emotional data and storing it in a database together with the feedback, and a means for retraining the artificial intelligence model based on the emotional data. This enables more accurate trend predictions that take the user's emotional state into account and personalized product recommendations for each user.

[0651] "Data collection means" refers to the ability to obtain information in real time from external data sources.

[0652] "Data formatting means" refers to the function of converting collected raw data into a usable format and storing it in a database.

[0653] "Feature extraction means" refers to the function that extracts important data points from stored data and uses them to train artificial intelligence models.

[0654] An "artificial intelligence model" refers to an algorithm or system that has been trained to make predictions or analyses.

[0655] "Trend Forecasting" refers to the ability to predict future trends using a trained artificial intelligence model.

[0656] "User interface" refers to a function or screen that visually presents prediction results to the user and allows them to operate them interactively.

[0657] "Feedback acquisition means" refers to the function of collecting opinions and advice from users and incorporating them into the system.

[0658] "Re-learning means" refers to the function of re-training the AI ​​model based on collected feedback to improve prediction accuracy.

[0659] "Emotional data" refers to information about a user's emotional state analyzed from their text and images.

[0660] "Emotional data collection means" refers to the function of analyzing the user's emotional state and storing it in a database.

[0661] The system for implementing the present invention consists of the following steps: data collection, data analysis and trend prediction, receiving and processing user feedback, collecting and integrating emotion data, and re-training. Each step is described in detail below.

[0662] Data collection

[0663] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. For example, news APIs, social media APIs, and economic indicator APIs are used. These raw data are converted into a data frame format and combined to create a unified dataset. This dataset is then stored in a database system, such as MySQL or PostgreSQL.

[0664] Feature extraction and AI model training

[0665] The server reads the stored data from the database and extracts the necessary features. Technically, this can be done using Python data analysis libraries such as pandas or numpy. Based on the extracted features, an artificial intelligence model is trained using a machine learning library such as scikit-learn. This AI model can use, for example, random forest or LSTM.

[0666] Future trend prediction

[0667] Using trained artificial intelligence models, the server predicts future trends. The results are then formatted into easy-to-understand visual displays (e.g., graphs and charts). For example, the next quarter's sales forecast is displayed in chart format.

[0668] Prediction results

[0669] When a user accesses the device, it sends a request to the server to get the latest trend forecast information. The server returns the forecast results to the device, which then displays them on a dashboard in the form of interactive graphs and charts. For example, the UI displays real-time stock price forecasts in infographic format.

[0670] Receiving and processing user feedback

[0671] Users access the system through their devices and input their predictions and opinions. This feedback is entered, for example, using an HTML form and sent to the server. The server receives it and stores it in a database. The stored feedback is then used to retrain the AI ​​model.

[0672] Collecting and processing emotional data

[0673] The server collects user emotion data through an emotion recognition engine. For example, libraries such as TextBlob and facial_emotion_recognition are used to analyze the emotion of facial expressions and text when users enter feedback. The resulting emotion data is then stored in a database along with the feedback.

[0674] Retraining using emotion data

[0675] The server retrains the AI ​​model based on the emotion data obtained from the emotion engine. This retraining process allows it to provide prediction results that take the user's emotional state into account, further improving prediction accuracy.

[0676] As a concrete example, the server retrieves business news from a news API, converts it into a data frame, and combines it. Features are extracted and an AI model is trained. Next, future trends are predicted and displayed in graph form. When a user accesses the system on their device, this predicted data is displayed on a dashboard. The user enters emotional data along with their own opinion, which is sent to the server and used for re-learning. An example of a prompt used here is, "Do you think sales will increase in the next quarter? Please explain why. Please also upload an image of your feedback."

[0677] The above is a specific embodiment of the present system. By adopting this system, companies can grasp market trends more precisely and in real time, enabling them to make effective decisions.

[0678] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0679] Step 1:

[0680] The server sends API requests from multiple data sources, such as news feeds, social media, and economic indicators, and retrieves data in real time.

[0681] Input: API request

[0682] Data processing: Convert the response from each API into a data frame format

[0683] Output: Data in data frame format

[0684] Step 2:

[0685] The server formats the acquired data into multiple data frames and combines them to create a unified dataset.

[0686] Input: Dataframe of news, social media, and economic indicator data

[0687] Data processing: Combining data frames

[0688] Output: Unified dataset

[0689] Step 3:

[0690] The server stores the unified data set in a database.

[0691] Input: Unified dataset

[0692] Action: Insert data into database

[0693] Output: Saved

[0694] Step 4:

[0695] The server reads the stored data from the database and extracts the necessary features.

[0696] Input: Data read from database

[0697] Data processing: feature extraction process

[0698] Output: Extracted features

[0699] Step 5:

[0700] The server trains an artificial intelligence model based on the extracted features.

[0701] Input: extracted features

[0702] Data processing: training artificial intelligence models

[0703] Output: A trained artificial intelligence model

[0704] Step 6:

[0705] The server uses trained artificial intelligence models to predict future trends.

[0706] Input: A trained AI model and the data to be predicted

[0707] Data calculation: Future trend prediction process

[0708] Output: Prediction results

[0709] Step 7:

[0710] The server formats the prediction results into interactive graphs and charts and sends them to the user interface.

[0711] Input: Prediction result

[0712] Data processing: Conversion into infographics format

[0713] Output: Interactive graphs and charts

[0714] Step 8:

[0715] The device displays the prediction results in the user interface and is ready to receive feedback from the user.

[0716] Input: Interactive graphs and charts

[0717] Behavior: Show in user interface

[0718] Output: Displayed prediction results

[0719] Step 9:

[0720] Users enter their predictions and opinions through a form and also upload image feedback.

[0721] Input: User opinions and feedback images

[0722] Action: Enter and send feedback

[0723] Output: Feedback data

[0724] Step 10:

[0725] The server performs text analysis on the received feedback to obtain emotion data.

[0726] Input: User feedback

[0727] Data Computing: Text Analysis and Emotion Recognition

[0728] Output: Emotion data

[0729] Step 11:

[0730] The server stores the acquired emotion data and feedback in a database.

[0731] Input: Emotion data and feedback

[0732] Action: Insert data into database

[0733] Output: Saved

[0734] Step 12:

[0735] The server retrains the artificial intelligence model based on the stored feedback and emotion data.

[0736] Input: Feedback and emotion data

[0737] Data Computing: Retraining Artificial Intelligence Models

[0738] Output: Updated artificial intelligence model

[0739] The above is the specific operation and data flow of each processing step in this system.

[0740] 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.

[0741] 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.

[0742] 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.

[0743] [Third embodiment]

[0744] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0745] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0746] 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).

[0747] 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.

[0748] 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.

[0749] 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).

[0750] 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.

[0751] 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.

[0752] 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.

[0753] 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.

[0754] 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.

[0755] 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."

[0756] A system for implementing the present invention combines a multi-step process that includes data collection, data analysis and trend prediction, receiving and processing user feedback, integrating expert insights, and presenting the predictive results.

[0757] Data collection

[0758] The server first collects data in real time. To do this, it obtains data from multiple data sources, such as news feeds, social media, and economic indicators, via APIs. The obtained data is then formatted and converted into a data frame. Multiple data frames are then combined to create a unified dataset. This unified dataset is then stored in a database. This process requires efficient data collection and storage.

[0759] Data analysis and trend forecasting

[0760] The server then reads the stored data from the database and extracts the necessary features. It then trains the data using an AI model, which uses algorithms such as random forest and LSTM. Once trained, the model is used to predict future trends. The prediction results are then formatted so that they are displayed to the user in a visually understandable format and sent to the user's device.

[0761] Receiving and processing feedback

[0762] Users access the system through their devices and input their predictions and opinions. For example, they can enter a prediction such as, "I think sales will increase next quarter." These predictions and opinions are sent to the server, which stores the received feedback in a database and further uses this feedback to update the training data and retrain the AI ​​model.

[0763] Integrating expert insights

[0764] Experts access the platform and input their own insights and analysis reports. These insights are sent to the server and stored in a database. The AI ​​model is then retrained based on the stored insights to improve its prediction accuracy. This process allows expert knowledge to be incorporated into the model.

[0765] Prediction results

[0766] When a user accesses the terminal, it sends a request to the server to obtain the latest trend forecast information. The server returns the latest forecast results, and the terminal displays these results on the dashboard in the form of interactive graphs and charts, allowing users to grasp the latest market trends at a glance.

[0767] As described above, the present invention provides a system that consistently performs everything from data collection to prediction and feedback integration, enabling companies to make more effective and faster decisions and helping them respond quickly to market changes.

[0768] The processing flow will be explained below.

[0769] Step 1:

[0770] Data collection and formatting (server)

[0771] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. The data is then formatted and converted into a data frame. These data frames are then combined to create a unified dataset. Finally, this unified dataset is stored in a database.

[0772] Step 2:

[0773] Feature extraction (server)

[0774] The server reads the stored data from the database and extracts the necessary features. Features are the data attributes and variables required to train a predictive model. These are important elements for correctly forecasting market trends.

[0775] Step 3:

[0776] AI model training (server)

[0777] The server trains an AI model based on the extracted features. Models used here include random forest and LSTM. The server uses these algorithms to learn the training data. Cross-validation and other processes are performed to improve the model's performance.

[0778] Step 4:

[0779] Future Trend Prediction (Server)

[0780] The server uses the trained AI model to predict future trends, and the prediction results are displayed in a visually easy-to-understand format (e.g., graphs and charts), allowing users to understand the prediction results at a glance.

[0781] Step 5:

[0782] Presenting prediction results (device)

[0783] When a user accesses the terminal, it sends a request to the server to obtain the latest trend forecast information. The server returns the forecast results to the terminal, which then displays them on the dashboard in the form of interactive graphs and charts, allowing users to understand market trends in real time.

[0784] Step 6:

[0785] Receiving user feedback (device, user)

[0786] Users input their predictions and opinions through their own devices. For example, they input predictions such as, "I think sales will increase in the next quarter." This feedback is sent to the server through a form.

[0787] Step 7:

[0788] Feedback storage and processing (server)

[0789] The server receives the feedback sent by the user and stores it in a database, after which the AI ​​model is retrained based on the stored feedback, a process that allows the accuracy of the prediction model to be improved.

[0790] Step 8:

[0791] Integration of expert insights (server, user)

[0792] Experts access the platform and input their own insights and analysis reports. The server stores these insights in a database and uses them to retrain the AI ​​model. By incorporating the experts' knowledge into the model, the accuracy of predictions can be further improved.

[0793] The above is the specific processing flow of this system.

[0794] Example 1

[0795] 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."

[0796] Traditional data analysis systems struggled to consistently collect data, forecast trends, and integrate feedback. This made it difficult for companies to obtain the latest market trend information in a timely manner to make quick decisions. Furthermore, there was a lack of a way to effectively integrate feedback and insights from users and experts to improve forecast accuracy.

[0797] 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.

[0798] In this invention, the server includes a data collection means, a means for formatting the collected data and storing it in a database, a means for extracting features from the stored data and training a machine learning model, a means for predicting future trends using the trained machine learning model, a means for sending the prediction results to a user interface and converting them into a displayable format, a means for obtaining feedback from users and retraining the machine learning model, a means for receiving expert insights through the platform and storing them in the database, and a means for retraining the machine learning model based on the stored expert insights to improve prediction accuracy. This enables companies to collect data in real time, make highly accurate trend predictions, and build a feedback loop that reflects the opinions of users and experts.

[0799] "Data collection means" refers to the means for obtaining data in real time from various data sources such as news feeds, social media, and economic indicators.

[0800] "Format" is the process of converting collected data into a usable format and carrying out the data cleaning and formatting required for storing it in a database.

[0801] "Means for storing in a database" refers to the means for storing formatted data in a database so that it can be efficiently searched and retrieved.

[0802] "Features" are attributes and values ​​of data extracted and transformed to be fed into a machine learning model, and contain information that is important for analysis and prediction.

[0803] A "machine learning model" is an algorithm or model that is trained using data to make predictions or classifications based on future data.

[0804] "Training means" refers to the means for training a machine learning model using collected data to improve the accuracy of the model.

[0805] A "means for predicting future trends" is a means for predicting trends in future data or events using a trained machine learning model.

[0806] A "user interface" is an interface through which a user accesses a system to input information or obtain information from the system.

[0807] The "means for converting into a displayable format" refers to a means for converting the prediction results into a visually easy-to-understand format (for example, a graph or chart) and displaying it on a user interface.

[0808] "Feedback" refers to predictions, opinions, and insights provided by users and experts.

[0809] "Means for obtaining feedback" refers to the means for collecting and storing feedback from users and experts.

[0810] "Retraining" refers to the process of retraining a machine learning model using feedback and new data to improve the model's accuracy.

[0811] "Platform" means a designated environment or system for experts to input and integrate their insights and analytical reports into the system.

[0812] "Insights" refers to detailed analysis and opinions provided by experts, and are important information for improving forecast accuracy.

[0813] The system of the present invention combines a multi-step process, including data collection, data analysis and trend prediction, receiving and processing user feedback, integrating expert insights, and presenting the prediction results. As a specific implementation method of the present invention, the following steps are described in detail.

[0814] Data collection

[0815] The server collects data in real time from multiple data sources, such as news feeds, social media, and economic indicators, via APIs. For example, the server uses Twitter's API to collect tweets based on specific keywords. It converts this data into a data frame using the Pandas library. The server then combines multiple data frames to create a unified dataset, which is then stored in an SQL database.

[0816] Data analysis and trend forecasting

[0817] The server reads the stored data from the database and extracts features. For example, the server extracts the number of tweets in the past 24 hours and the percentage of tweets with positive context. The server then trains a machine learning model such as random forest or LSTM using Scikit-learn or TensorFlow. The trained model is used to predict future trends. The prediction results are formatted into a visually easy-to-understand format (e.g., graphs or charts) using the Plotly library and sent to the user's device.

[0818] Receiving and processing feedback

[0819] Users use their own devices to input their predictions and opinions into the system. For example, a user can input a prediction such as, "I think sales will increase next quarter." The device then sends the input feedback to the server. The server stores the received feedback in a database and updates the training data based on the stored feedback. The server then retrains the machine learning model using the updated data.

[0820] Integrating expert insights

[0821] Experts access the platform and input their own insights and analysis reports. The device then sends these insights to the server, which stores them in a database and retrains the machine learning model based on the stored insights. This incorporates expert knowledge into the model, improving its prediction accuracy.

[0822] Prediction results

[0823] The user's device sends a request to the server to get the latest trend forecast information, and the server returns the latest forecast results, which the device displays on a dashboard in the form of interactive graphs and charts.

[0824] Specific examples

[0825] For example, a user can enter the following prompt into the system:

[0826] "Collect the latest tweet data and make supply and demand forecasts."

[0827] "Create a sales forecast for the next quarter and display it in a chart."

[0828] "Update the AI ​​model based on user feedback."

[0829] In response to these prompts, the server collects, analyzes, and processes appropriate data. Furthermore, by incorporating expert insights into the model, forecast accuracy can be improved. In this way, the system of the present invention helps companies respond quickly to market changes and make effective decisions.

[0830] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0831] Step 1: Data collection

[0832] The server uses APIs to collect data in real time from various data sources, such as news feeds, social media, and economic indicators. For example, the Twitter API is used to collect tweets based on specific keywords. The server periodically executes scheduled tasks to obtain the data. The collected data is converted into a data frame using the Pandas library. The input is raw data from each data source, and the output is a formatted data frame.

[0833] Step 2: Format and save the data

[0834] The server reformats the collected data using the Pandas library. This reformatting process includes removing duplicate data, handling missing values, and normalizing text. The reformatted data is then combined to create a unified dataset. This dataset is then stored in a SQL database. The input is an unformed data frame, and the output is the reformatted and combined dataset.

[0835] Step 3: Feature extraction

[0836] The server reads the stored data from the database and extracts features, such as the number of tweets per hour or the percentage of tweets with positive context. The Scikit-learn library is used to encode these features. The input is the dataset, and the output is a list of numerically encoded features.

[0837] Step 4: Train the model

[0838] The server trains the extracted features using machine learning models such as random forests and LSTM. For example, it uses past tweet data to predict future trends. Libraries such as TensorFlow and Scikit-learn are used for this training. The accuracy of the trained model is verified and saved. The input is a list of features, and the output is a trained machine learning model.

[0839] Step 5: Predicting trends

[0840] The server uses the trained model to predict future trends, for example, forecasting sales for the next quarter. The prediction results are visualized using the Plotly library and output in a format that is easy for users to understand. The input is the machine learning model and features, and the output is the visualized prediction results.

[0841] Step 6: Receiving feedback

[0842] Users input their predictions and opinions into the system through their terminals. For example, they input a prediction such as, "I think sales will increase next quarter." The terminals collect this feedback and send it to the server. The input is the user's feedback, and the output is the feedback data sent to the server.

[0843] Step 7: Process feedback and retrain

[0844] The server stores the received feedback in a database and updates the training data based on the stored feedback. The AI ​​model is retrained using the updated data, which improves the model's accuracy. The input is the feedback data, and the output is an updated machine learning model.

[0845] Step 8: Integrating expert insights

[0846] Experts access the platform and input their insights and analysis reports. The device sends them to the server, which stores the received insights in a database. The stored insights are used to retrain the machine learning model, further improving its prediction accuracy. The input is the expert's insights, and the output is the retrained machine learning model.

[0847] Step 9: Presenting the prediction results

[0848] The user's device requests the latest forecast results from the server. The server returns the latest forecast results, and the device displays these results on a dashboard as interactive graphs and charts, allowing the user to understand the latest market trends at a glance. The input is a request for forecast results, and the output is the forecast results in a displayable format.

[0849] (Application example 1)

[0850] 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."

[0851] Conventional ad delivery systems lacked real-time market trend analysis, making effective targeting and personalization difficult. They also struggled to quickly incorporate user and expert feedback, delaying the optimization of ad campaigns. Furthermore, limited means of collecting and analyzing user behavior data made it difficult to accurately evaluate advertising effectiveness.

[0852] 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.

[0853] In this invention, the server includes a data collection means, a means for formatting the collected data and storing it in a database, a means for extracting features from the stored data and training an AI model, a means for predicting future trends using the trained AI model, a means for sending the prediction results to a user interface and converting them into a displayable format, a means for receiving feedback from users and retraining the AI ​​model, a means for receiving insights and analysis reports from experts and storing them in a database and retraining the AI ​​model, and a means for generating personalized advertisements based on the collected and analyzed data and displaying them on user terminals. This enables effective targeting based on real-time market trend analysis, optimization of advertising campaigns that quickly reflect user and expert feedback, and accurate evaluation of advertising effectiveness by collecting and analyzing user behavior data.

[0854] "Data collection means" refers to a means for obtaining data in real time from multiple data sources such as news feeds, social media, and economic indicators.

[0855] "Database" means a centralized data repository for storing uniform data sets for later use in analysis and prediction.

[0856] A "feature extraction method" is a method for identifying information necessary for trend prediction from collected data and converting it into a format that can be used to train an AI model.

[0857] "AI model training method" refers to a method of training an AI model (such as random forest or LSTM) using extracted features to enable it to predict future trends.

[0858] A "trend forecasting tool" is a tool for forecasting future trends using a trained AI model.

[0859] The "user interface transmission means" is a means for converting the prediction results into a format that is easy for the user to understand and transmitting it to the user's terminal.

[0860] "Feedback acquisition means" refers to the means of collecting predictions and opinions from users and reflecting them in the system.

[0861] "AI model retraining method" refers to a method of retraining an AI model based on feedback from users and experts to improve its accuracy.

[0862] The "means for receiving expert insights" is a means for receiving insights and analysis reports from experts, storing them in a database, and then retraining the AI ​​model.

[0863] "Personalized advertisement generation means" refers to a means for generating advertisements optimized for individual users based on collected and analyzed data and displaying them on the user's terminal.

[0864] The system that realizes this application example comprises the smartphone app "AdTrend." The server collects data in real time from multiple data sources, such as news feeds, social media, and economic indicators. The collected data is formatted and converted into a data frame, and these data frames are then combined to create a unified dataset. This process uses Python's Pandas framework.

[0865] Next, the server reads the stored data from the database and extracts the necessary features. Based on the extracted features, it trains an AI model using algorithms such as random forest and LSTM. This training generates an AI model that can predict future trends. The model is trained using the Scikit-learn library and generates prediction results.

[0866] The server then formats these forecasts in JSON format and sends them to the user's smartphone, where they are displayed as interactive charts and graphs. This dashboard helps users visually understand trends.

[0867] Users can enter their predictions and opinions into the system through a form. This feedback is sent to the server and stored in a database. The stored feedback is used as new training data to retrain the AI ​​model. The Flask framework is used to receive and process user feedback.

[0868] In addition, insights and analysis reports from experts are also sent to the server and stored in the database, allowing the experts' knowledge to be utilized to improve the prediction accuracy of the AI ​​model. In this process, the experts' insights are incorporated into the retraining of the AI ​​model.

[0869] Regarding the generation of personalized ads, the collected data is used to generate ads optimized for individual user attributes. These ads are delivered and displayed on the user's device. User behavior data is collected and analyzed to evaluate the effectiveness of the ads. Ad campaigns are optimized using Python and related libraries.

[0870] Examples of specific prompts include:

[0871] Trend Forecast:

[0872] Analyze the characteristics of the most popular advertising campaigns in the next quarter and propose effective advertising strategies.

[0873] Feedback received:

[0874] We collect user opinions about the ads currently being served and use that data to personalize ads.

[0875] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0876] Step 1:

[0877] The server collects data in real time from multiple data sources, such as news feeds, social media, and economic indicators. It uses an API to retrieve the data and converts it into a Pandas data frame. The input is raw data from each data source, and the output is organized data in a data frame format.

[0878] Step 2:

[0879] The server combines the multiple formatted data frames to create a unified dataset and stores it in the database. The input is multiple data frames, and the output is the combined unified dataset. Specifically, the server combines the data frames using functions such as the merge function in Pandas and stores it in the database.

[0880] Step 3:

[0881] The server reads the stored data from the database and extracts the necessary features. The input is a unified dataset, and the output is the features required for training the AI ​​model. Specifically, it uses Pandas and Scikit-learn to extract the features and prepare them as training data.

[0882] Step 4:

[0883] The server uses the extracted features to train an AI model. The input is the features, and the output is a trained AI model (e.g., random forest or LSTM). Specifically, the server uses the Scikit-learn library to train the model and check its prediction accuracy.

[0884] Step 5:

[0885] The server uses a trained AI model to predict future trends. The input is newly collected data, and the output is the future trend prediction result. Specifically, new data is input into the model and a prediction result is generated.

[0886] Step 6:

[0887] The server formats the prediction results in JSON format and sends them to the user's device. The input is the prediction result data, and the output is the JSON-formatted prediction result sent to the user's device. Specifically, Flask is used to communicate data between the server and the device.

[0888] Step 7:

[0889] The terminal displays the received prediction results as interactive charts and graphs. The input is the prediction result data in JSON format, and the output is the prediction result displayed graphically. Specifically, a dashboard is created using a JavaScript library (e.g., D3.js) to provide visual feedback to the user.

[0890] Step 8:

[0891] Users enter their predictions and opinions into the system through a form. The input is the user's feedback, and the output is the feedback data sent to the server. Specifically, the system uses a combination of HTML forms and JavaScript to send user input to the server.

[0892] Step 9:

[0893] The server receives feedback from the user and stores it in a database. The input is the feedback data, and the output is the feedback stored in the database. Specifically, Flask is used to receive the data and perform the storage process.

[0894] Step 10:

[0895] The server then retrains the AI ​​model using the stored feedback. The input is the feedback data in the database, and the output is the retrained AI model. Specifically, it repeats feature extraction and model training as before.

[0896] Step 11:

[0897] The expert inputs his / her insights and analysis reports and sends them to the server. The input is the expert's insights and analysis reports, and the output is the insight data sent to the server. The specific operation is to accept the expert's input via a form.

[0898] Step 12:

[0899] The server stores the insights received from the experts in a database and uses them to retrain the AI ​​model. The input is the expert's insight data, and the output is a retrained AI model that reflects the insights. Specifically, the insight data is stored in a database and used for the re-learning process.

[0900] Step 13:

[0901] The server generates personalized advertisements based on the user's attribute data and displays them on the user's device. The input is the user's attribute data, and the output is personalized advertisements. Specifically, the advertisement content is dynamically changed based on the attribute data and delivered to the user.

[0902] Step 14:

[0903] The server collects user behavior data and evaluates the effectiveness of advertising. The input is user behavior data, and the output is evaluated advertising effectiveness data. Specific operations include analyzing the behavior data and calculating evaluation indicators.

[0904] 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.

[0905] The system for implementing the present invention consists of a multi-step process that combines data collection, data analysis and trend prediction, receiving and processing user feedback, integrating expert insights, presenting prediction results, and an emotion engine that recognizes user emotions.

[0906] Data collection and formatting

[0907] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. It formats this data and converts it into a data frame format. It combines multiple data frames to create a unified dataset and stores this dataset in a database. As an example, it retrieves the latest business news from a news API and converts it into a data frame.

[0908] Feature extraction and AI model training

[0909] The server reads the stored data from the database and extracts the necessary features. An AI model is trained based on the extracted features. This AI model uses algorithms such as random forest and LSTM. Features include market prices, sales data, and seasonal factors.

[0910] Future trend prediction

[0911] Using the trained AI model, the server predicts future trends. The prediction results are displayed in a visually easy-to-understand format (for example, graphs or charts). This allows users to understand the prediction results at a glance. For example, the sales forecast for the next quarter is displayed in chart format.

[0912] Prediction results

[0913] When a user accesses the device, it sends a request to the server to get the latest trend forecast information. The server returns the forecast results to the device, which then displays them on a dashboard in the form of interactive graphs and charts. For example, the UI displays real-time stock price forecasts in infographic format.

[0914] Receiving and processing user feedback

[0915] Users access the system through their own devices and input their predictions and opinions. This feedback is sent to the server and stored in a database. The stored feedback is then used to retrain the AI ​​model. For example, a user's opinion such as "the next sales will increase" is incorporated as retraining data.

[0916] Integrating expert insights

[0917] Experts access the platform and input their own insights and analysis reports, which are then sent to the server and stored in a database.The AI ​​model is then retrained based on the stored insights to improve its prediction accuracy.

[0918] Collecting and processing user sentiment data

[0919] The server collects user emotional data through the emotion engine. For example, when a user enters feedback, the server analyzes emotions from their facial expressions and text to obtain emotional data. This emotional data is then stored in a database along with the user's feedback and opinions.

[0920] Retraining using emotion data

[0921] The server retrains the AI ​​model based on the emotional data obtained from the emotion engine. This retraining process makes it possible to provide prediction results that take the user's emotional state into account, further improving prediction accuracy.

[0922] Specific examples

[0923] For example, the server retrieves business news from a news API, converts it into a data frame, and combines it. Features are extracted and used to train an AI model. Future trends are then predicted and displayed in graph form. When a user accesses the system on their device, this predicted data is displayed on a dashboard. Users input their own opinions and sentiment data, which is sent to the server and used for retraining. Experts also provide insights, improving the overall accuracy of the AI ​​model.

[0924] This is the specific processing flow of this system. By adopting this system, companies will be able to grasp market trends more precisely and in real time, enabling them to make effective decisions.

[0925] The processing flow will be explained below.

[0926] Step 1:

[0927] Data collection and formatting (server)

[0928] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. The retrieved data is converted from JSON format to a data frame format. The server then combines multiple data frames to create a unified dataset and stores this dataset in a database. For example, data retrieved from an economic indicator API is updated daily and stored in a database.

[0929] Step 2:

[0930] Feature extraction (server)

[0931] The server reads the stored data from the database and extracts the necessary features. These features range from sales data, consumer sentiment trends, seasonal factors, product ratings, etc. These features are necessary for the AI ​​model to accurately predict trends.

[0932] Step 3:

[0933] AI model training (server)

[0934] The server trains an AI model based on the extracted features. This AI model uses algorithms such as random forest and LSTM. It uses daily sales data and economic indicators as features and learns the importance of each variable.

[0935] Step 4:

[0936] Future Trend Prediction (Server)

[0937] The server uses the trained AI model to predict future trends. The prediction results are displayed in a visually easy-to-understand format (e.g., graphs and charts), allowing users to understand the prediction results at a glance.

[0938] Step 5:

[0939] Presenting prediction results (device)

[0940] When a user accesses the device, it sends a request to the server to get the latest trend forecast information. The server then sends the forecast results to the device, which then displays them on a dashboard in the form of interactive graphs and charts. For example, the next quarter's sales forecast can be displayed in infographic format.

[0941] Step 6:

[0942] Receiving user feedback (user, device)

[0943] Users access the system through their own terminals and input their predictions and opinions. For example, they can input a prediction such as, "I think sales will increase next quarter." This feedback is sent to the server through a form.

[0944] Step 7:

[0945] Feedback storage and processing (server)

[0946] The server receives the feedback sent by the user and stores it in a database. It then retrains the AI ​​model based on the stored feedback. This retraining process improves the accuracy of the prediction model.

[0947] Step 8:

[0948] Integration of expert insights (server, user)

[0949] Experts access the platform and input their own insights and analytical reports, which the server then stores in a database and uses to retrain the AI ​​model, incorporating the experts' knowledge to further improve its prediction accuracy.

[0950] Step 9:

[0951] Collecting user emotion data (device, user)

[0952] When the user inputs feedback, the device collects the user's emotional data through the emotion engine. For example, when the user inputs feedback, the device analyzes emotions from the user's facial expressions and text to obtain the emotional data.

[0953] Step 10:

[0954] Storage and processing of emotional data (server)

[0955] The server receives the emotion data sent from the device and stores it in a database along with the feedback data. The emotion data is then used to retrain the AI ​​model. This process allows it to provide prediction results that take the user's emotional state into account.

[0956] This is the specific processing flow of this system, which enables companies to make accurate predictions based on multifaceted data, including user sentiment, and supports more effective and faster decision-making.

[0957] Example 2

[0958] 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."

[0959] In today's business environment, accurately predicting market trends is extremely difficult, and companies require real-time analysis and forecasting to make effective decisions. Traditional forecasting systems struggle to take user feedback and sentiment into account, resulting in poor forecast accuracy. Furthermore, the lack of a means to effectively integrate expert insights limits overall forecast accuracy.

[0960] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a data collection means, a means for formatting the collected data and storing it in an information storage device, a means for extracting features based on the stored data and training a generative learning model, a means for predicting future trends using the trained generative learning model, a means for transmitting the prediction results to a terminal device interface and converting them into a displayable format, a means for obtaining feedback from users and retraining the generative learning model, a means for collecting user emotion data using an emotion recognition device, and a means for retraining the generative learning model based on the collected emotion data. This makes it possible to collect and format information from multiple data sources in real time and accurately predict future trends using the generative learning model. Furthermore, taking user feedback and emotions into account improves prediction accuracy, and integrating it with expert insights can further improve prediction accuracy.

[0961] "Data collection means" refers to a means for obtaining data in real time from multiple data sources such as news feeds, social media, and economic indicators.

[0962] An "information storage device" is a database or storage system for saving and managing formatted data.

[0963] "Features" are key data elements extracted from stored data and used to train predictive models.

[0964] A "generative learning model" is an AI model that learns from training data and predicts future trends.

[0965] "Endpoint interface" refers to the interactive display screen or dashboard that a user uses to view prediction results.

[0966] "User feedback" refers to predictions and opinions entered by users through the system.

[0967] "Retraining" is the process of adding new data to an existing training model to improve the model's predictive accuracy.

[0968] An "emotion recognition device" is an engine or system that analyzes and acquires emotional data from user feedback, facial expressions, and text.

[0969] "Emotional data" refers to data that represents a user's emotional state, including emotional categories such as positive, negative, and neutral.

[0970] "Trend forecasting" refers to the process of using trained generative learning models to predict future market movements and trends.

[0971] The present invention is a system for forecasting future market trends by collecting data from multiple data sources in real time, training and relearning a generative learning model based on the collected data, and integrating feedback from servers, terminals, and users to improve forecast accuracy.

[0972] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. It uses common data processing libraries (e.g., Pandas, NumPy) to format this data and convert it into a data frame. It then combines these data frames to create a unified dataset and stores it in a database (e.g., MySQL, PostgreSQL).

[0973] Next, the server reads the stored data from the database and extracts the necessary features. This extraction uses a data analysis library (e.g., Scikit-learn, TensorFlow). Based on the extracted features, a generative learning model (e.g., Random Forest, LSTM) is trained. This training process allows the AI ​​model to learn trends from past data and predict future trends.

[0974] Using the trained generative learning model, the server predicts future trends and formats the prediction results in a visually understandable format such as graphs and charts. This visualization is done using visualization libraries (e.g., Matplotlib, Plotly). The prediction results are sent to the end device interface and displayed on an interactive dashboard when accessed by the user.

[0975] Users access the system through their devices and input their predictions and opinions. This feedback is sent to the server and stored in a database. Based on the stored feedback, the generative learning model is retrained to improve prediction accuracy. This retraining process uses techniques to update the existing model using new data.

[0976] The system also includes an emotion recognizer that collects emotion data from user feedback, facial expressions, and text. Emotion recognition uses natural language processing libraries (e.g., NLTK, spaCy) and face recognition libraries (e.g., OpenCV). The collected emotion data is stored in a database and used to retrain a generative learning model. This allows prediction results to be provided that take the user's emotional state into account.

[0977] As a concrete example, the server retrieves the latest business news from a news API, converts the article contents into a data frame, and combines them. It then extracts features from market prices and performance data to train a generative learning model. It predicts sales for the next quarter and displays the results in a graph format. When accessed by a user, the predicted data is displayed on the device's dashboard, and the user can enter their own feedback and sentiment data. This is sent to the server and used for re-training. Experts also provide insights, improving the accuracy of the generative learning model.

[0978] Prompt Sentence Examples

[0979] "Please use the Random Forest algorithm to predict sales for the next quarter and display the results in a graph."

[0980] "Predict market trends based on sentiment data and present the results to users in infographics."

[0981] "Retrain your generative learning model with user feedback to improve prediction accuracy."

[0982] By adopting this system, companies will be able to collect information from multiple data sources in real time, integrate the opinions of users and experts, and gain a more precise understanding of market trends in real time, enabling them to make more effective decisions.

[0983] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0984] System program processing flow

[0985] Step 1: Data collection

[0986] The server retrieves data in real time from multiple data sources, such as news feeds, social media, and economic indicators. Specifically, it sends API requests to retrieve news articles, social media posts, and economic indicator data. As input, it uses API endpoints and access keys. As output, it returns raw data to the server.

[0987] Step 2: Data Formatting and Storage

[0988] The server formats the retrieved data and converts it into a data frame. Specifically, it extracts information such as the title, content, and date and time of the retrieved news articles and formats them into a data frame using the Pandas or NumPy library. Raw data is used as input, and a data frame is obtained as output. This is then saved in a database. For example, it connects to a database such as MySQL or PostgreSQL and inserts the data.

[0989] Step 3: Feature extraction

[0990] The server reads the stored data from the database and extracts the necessary features. Specifically, features such as market prices, sales data, and seasonal factors are calculated using scripts and extracted using data analysis libraries (e.g., Scikit-learn, TensorFlow). The data read from the database is used as input, and the features are obtained as output.

[0991] Step 4: Training the AI ​​model

[0992] The server trains a generative AI model based on the extracted features. Algorithms used include random forest and LSTM. Specifically, the model is trained using a training dataset and the trained model is saved. The feature dataset is used as input, and the trained model is obtained as output.

[0993] Step 5: Trend forecasting

[0994] The server uses a trained generative AI model to predict future trends. Specifically, it inputs new data into the trained model to predict future market trends and sales. New data is used as input, and predictions are obtained as output. The predictions are then formatted in a visually easy-to-understand format (e.g., graphs or charts).

[0995] Step 6: Formatting and presenting prediction results

[0996] When a user accesses the device, it sends a request to obtain the latest trend forecast information from the server. The server returns the forecast results to the device, which then displays them on a dashboard in the form of interactive graphs and charts. For example, the forecast results can be visualized using libraries such as Plotly or D3.js. The forecast results are used as input, and a visualized dashboard is obtained as output.

[0997] Step 7: Collect and process user feedback

[0998] Users access the system through their own devices and enter their predictions and opinions. A feedback form is provided as input, and users enter their opinions and predictions. This feedback is sent to the server and stored in a database. The saved feedback data is obtained as output.

[0999] Step 8: Integrating expert insights

[1000] Experts access the platform and input their own insights and analytical reports. These insights are sent to the server and stored in the database. The expert insights are used as input and the stored insight data is obtained as output. The generative AI model is then retrained based on the stored insights to improve prediction accuracy.

[1001] Step 9: Collect and process emotion data

[1002] The server uses an emotion recognition device to collect user emotion data. Specifically, when the user enters feedback, it analyzes emotions from their facial expressions and text. It uses libraries such as OpenCV and NLTK. It uses the user's feedback data and facial expression data as input, and obtains analyzed emotion data as output. This is then stored in a database.

[1003] Step 10: Retraining with emotion data

[1004] The server retrains the generative AI model based on the collected emotional data. Specifically, the emotional data is incorporated into the existing learning model as features to improve prediction accuracy. The emotional data and feature data are used as input, and a retrained generative AI model is obtained as output. This provides prediction results that take the user's emotional state into account.

[1005] (Application example 2)

[1006] 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."

[1007] Conventional trend prediction systems have difficulty taking user feedback into account, particularly the influence of user emotions. Furthermore, there is a lack of a means to efficiently integrate real-time user feedback and expert insights, limiting the accuracy of predictions. This makes it difficult to provide product recommendations that meet user needs, especially on online shopping sites.

[1008] 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.

[1009] In this invention, the server includes a data collection means, a means for formatting the collected data and storing it in a database, a means for extracting features from the stored data and training an artificial intelligence model, a means for predicting future trends using the trained artificial intelligence model, a means for sending the prediction results to a user interface and converting them into a displayable format, a means for obtaining feedback from users and retraining the artificial intelligence model, a means for collecting user emotional data and storing it in a database together with the feedback, and a means for retraining the artificial intelligence model based on the emotional data. This enables more accurate trend predictions that take the user's emotional state into account and personalized product recommendations for each user.

[1010] "Data collection means" refers to the ability to obtain information in real time from external data sources.

[1011] "Data formatting means" refers to the function of converting collected raw data into a usable format and storing it in a database.

[1012] "Feature extraction means" refers to the function that extracts important data points from stored data and uses them to train artificial intelligence models.

[1013] An "artificial intelligence model" refers to an algorithm or system that has been trained to make predictions or analyses.

[1014] "Trend Forecasting" refers to the ability to predict future trends using a trained artificial intelligence model.

[1015] "User interface" refers to a function or screen that visually presents prediction results to the user and allows them to operate them interactively.

[1016] "Feedback acquisition means" refers to the function of collecting opinions and advice from users and incorporating them into the system.

[1017] "Re-learning means" refers to the function of re-training the AI ​​model based on collected feedback to improve prediction accuracy.

[1018] "Emotional data" refers to information about a user's emotional state analyzed from their text and images.

[1019] "Emotional data collection means" refers to the function of analyzing the user's emotional state and storing it in a database.

[1020] The system for implementing the present invention consists of the following steps: data collection, data analysis and trend prediction, receiving and processing user feedback, collecting and integrating emotion data, and re-training. Each step is described in detail below.

[1021] Data collection

[1022] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. For example, news APIs, social media APIs, and economic indicator APIs are used. These raw data are converted into a data frame format and combined to create a unified dataset. This dataset is then stored in a database system, such as MySQL or PostgreSQL.

[1023] Feature extraction and AI model training

[1024] The server reads the stored data from the database and extracts the necessary features. Technically, this can be done using Python data analysis libraries such as pandas or numpy. Based on the extracted features, an artificial intelligence model is trained using a machine learning library such as scikit-learn. This AI model can use, for example, random forest or LSTM.

[1025] Future trend prediction

[1026] Using trained artificial intelligence models, the server predicts future trends. The results are then formatted into easy-to-understand visual displays (e.g., graphs and charts). For example, the next quarter's sales forecast is displayed in chart format.

[1027] Prediction results

[1028] When a user accesses the device, it sends a request to the server to get the latest trend forecast information. The server returns the forecast results to the device, which then displays them on a dashboard in the form of interactive graphs and charts. For example, the UI displays real-time stock price forecasts in infographic format.

[1029] Receiving and processing user feedback

[1030] Users access the system through their devices and input their predictions and opinions. This feedback is entered, for example, using an HTML form and sent to the server. The server receives it and stores it in a database. The stored feedback is then used to retrain the AI ​​model.

[1031] Collecting and processing emotional data

[1032] The server collects user emotion data through an emotion recognition engine. For example, libraries such as TextBlob and facial_emotion_recognition are used to analyze the emotion of facial expressions and text when users enter feedback. The resulting emotion data is then stored in a database along with the feedback.

[1033] Retraining using emotion data

[1034] The server retrains the AI ​​model based on the emotion data obtained from the emotion engine. This retraining process allows it to provide prediction results that take the user's emotional state into account, further improving prediction accuracy.

[1035] As a concrete example, the server retrieves business news from a news API, converts it into a data frame, and combines it. Features are extracted and an AI model is trained. Next, future trends are predicted and displayed in graph form. When a user accesses the system on their device, this predicted data is displayed on a dashboard. The user enters emotional data along with their own opinion, which is sent to the server and used for re-learning. An example of a prompt used here is, "Do you think sales will increase in the next quarter? Please explain why. Please also upload an image of your feedback."

[1036] The above is a specific embodiment of the present system. By adopting this system, companies can grasp market trends more precisely and in real time, enabling them to make effective decisions.

[1037] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1038] Step 1:

[1039] The server sends API requests from multiple data sources, such as news feeds, social media, and economic indicators, and retrieves data in real time.

[1040] Input: API request

[1041] Data processing: Convert the response from each API into a data frame format

[1042] Output: Data in data frame format

[1043] Step 2:

[1044] The server formats the acquired data into multiple data frames and combines them to create a unified dataset.

[1045] Input: Dataframe of news, social media, and economic indicator data

[1046] Data processing: Combining data frames

[1047] Output: Unified dataset

[1048] Step 3:

[1049] The server stores the unified data set in a database.

[1050] Input: Unified dataset

[1051] Action: Insert data into database

[1052] Output: Saved

[1053] Step 4:

[1054] The server reads the stored data from the database and extracts the necessary features.

[1055] Input: Data read from database

[1056] Data processing: feature extraction process

[1057] Output: Extracted features

[1058] Step 5:

[1059] The server trains an artificial intelligence model based on the extracted features.

[1060] Input: extracted features

[1061] Data processing: training artificial intelligence models

[1062] Output: A trained artificial intelligence model

[1063] Step 6:

[1064] The server uses trained artificial intelligence models to predict future trends.

[1065] Input: A trained AI model and the data to be predicted

[1066] Data calculation: Future trend prediction process

[1067] Output: Prediction results

[1068] Step 7:

[1069] The server formats the prediction results into interactive graphs and charts and sends them to the user interface.

[1070] Input: Prediction result

[1071] Data processing: Conversion into infographics format

[1072] Output: Interactive graphs and charts

[1073] Step 8:

[1074] The device displays the prediction results in the user interface and is ready to receive feedback from the user.

[1075] Input: Interactive graphs and charts

[1076] Behavior: Show in user interface

[1077] Output: Displayed prediction results

[1078] Step 9:

[1079] Users enter their predictions and opinions through a form and also upload image feedback.

[1080] Input: User opinions and feedback images

[1081] Action: Enter and send feedback

[1082] Output: Feedback data

[1083] Step 10:

[1084] The server performs text analysis on the received feedback to obtain emotion data.

[1085] Input: User feedback

[1086] Data Computing: Text Analysis and Emotion Recognition

[1087] Output: Emotion data

[1088] Step 11:

[1089] The server stores the acquired emotion data and feedback in a database.

[1090] Input: Emotion data and feedback

[1091] Action: Insert data into database

[1092] Output: Saved

[1093] Step 12:

[1094] The server retrains the artificial intelligence model based on the stored feedback and emotion data.

[1095] Input: Feedback and emotion data

[1096] Data Computing: Retraining Artificial Intelligence Models

[1097] Output: Updated artificial intelligence model

[1098] The above is the specific operation and data flow of each processing step in this system.

[1099] 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.

[1100] 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.

[1101] 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.

[1102] [Fourth embodiment]

[1103] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1104] 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.

[1105] 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).

[1106] 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.

[1107] 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.

[1108] 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).

[1109] 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.

[1110] 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.

[1111] 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.

[1112] 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.

[1113] 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.

[1114] 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.

[1115] 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."

[1116] A system for implementing the present invention combines a multi-step process that includes data collection, data analysis and trend prediction, receiving and processing user feedback, integrating expert insights, and presenting the predictive results.

[1117] Data collection

[1118] The server first collects data in real time. To do this, it obtains data from multiple data sources, such as news feeds, social media, and economic indicators, via APIs. The obtained data is then formatted and converted into a data frame. Multiple data frames are then combined to create a unified dataset. This unified dataset is then stored in a database. This process requires efficient data collection and storage.

[1119] Data analysis and trend forecasting

[1120] The server then reads the stored data from the database and extracts the necessary features. It then trains the data using an AI model, which uses algorithms such as random forest and LSTM. Once trained, the model is used to predict future trends. The prediction results are then formatted so that they are displayed to the user in a visually understandable format and sent to the user's device.

[1121] Receiving and processing feedback

[1122] Users access the system through their devices and input their predictions and opinions. For example, they can enter a prediction such as, "I think sales will increase next quarter." These predictions and opinions are sent to the server, which stores the received feedback in a database and further uses this feedback to update the training data and retrain the AI ​​model.

[1123] Integrating expert insights

[1124] Experts access the platform and input their own insights and analysis reports. These insights are sent to the server and stored in a database. The AI ​​model is then retrained based on the stored insights to improve its prediction accuracy. This process allows expert knowledge to be incorporated into the model.

[1125] Prediction results

[1126] When a user accesses the terminal, it sends a request to the server to obtain the latest trend forecast information. The server returns the latest forecast results, and the terminal displays these results on the dashboard in the form of interactive graphs and charts, allowing users to grasp the latest market trends at a glance.

[1127] As described above, the present invention provides a system that consistently performs everything from data collection to prediction and feedback integration, enabling companies to make more effective and faster decisions and helping them respond quickly to market changes.

[1128] The processing flow will be explained below.

[1129] Step 1:

[1130] Data collection and formatting (server)

[1131] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. The data is then formatted and converted into a data frame. These data frames are then combined to create a unified dataset. Finally, this unified dataset is stored in a database.

[1132] Step 2:

[1133] Feature extraction (server)

[1134] The server reads the stored data from the database and extracts the necessary features. Features are the data attributes and variables required to train a predictive model. These are important elements for correctly forecasting market trends.

[1135] Step 3:

[1136] AI model training (server)

[1137] The server trains an AI model based on the extracted features. Models used here include random forest and LSTM. The server uses these algorithms to learn the training data. Cross-validation and other processes are performed to improve the model's performance.

[1138] Step 4:

[1139] Future Trend Prediction (Server)

[1140] The server uses the trained AI model to predict future trends, and the prediction results are displayed in a visually easy-to-understand format (e.g., graphs and charts), allowing users to understand the prediction results at a glance.

[1141] Step 5:

[1142] Presenting prediction results (device)

[1143] When a user accesses the terminal, it sends a request to the server to obtain the latest trend forecast information. The server returns the forecast results to the terminal, which then displays them on the dashboard in the form of interactive graphs and charts, allowing users to understand market trends in real time.

[1144] Step 6:

[1145] Receiving user feedback (device, user)

[1146] Users input their predictions and opinions through their own devices. For example, they input predictions such as, "I think sales will increase in the next quarter." This feedback is sent to the server through a form.

[1147] Step 7:

[1148] Feedback storage and processing (server)

[1149] The server receives the feedback sent by the user and stores it in a database, after which the AI ​​model is retrained based on the stored feedback, a process that allows the accuracy of the prediction model to be improved.

[1150] Step 8:

[1151] Integration of expert insights (server, user)

[1152] Experts access the platform and input their own insights and analysis reports. The server stores these insights in a database and uses them to retrain the AI ​​model. By incorporating the experts' knowledge into the model, the accuracy of predictions can be further improved.

[1153] The above is the specific processing flow of this system.

[1154] Example 1

[1155] 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."

[1156] Traditional data analysis systems struggled to consistently collect data, forecast trends, and integrate feedback. This made it difficult for companies to obtain the latest market trend information in a timely manner to make quick decisions. Furthermore, there was a lack of a way to effectively integrate feedback and insights from users and experts to improve forecast accuracy.

[1157] 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.

[1158] In this invention, the server includes a data collection means, a means for formatting the collected data and storing it in a database, a means for extracting features from the stored data and training a machine learning model, a means for predicting future trends using the trained machine learning model, a means for sending the prediction results to a user interface and converting them into a displayable format, a means for obtaining feedback from users and retraining the machine learning model, a means for receiving expert insights through the platform and storing them in the database, and a means for retraining the machine learning model based on the stored expert insights to improve prediction accuracy. This enables companies to collect data in real time, make highly accurate trend predictions, and build a feedback loop that reflects the opinions of users and experts.

[1159] "Data collection means" refers to the means for obtaining data in real time from various data sources such as news feeds, social media, and economic indicators.

[1160] "Format" is the process of converting collected data into a usable format and carrying out the data cleaning and formatting required for storing it in a database.

[1161] "Means for storing in a database" refers to the means for storing formatted data in a database so that it can be efficiently searched and retrieved.

[1162] "Features" are attributes and values ​​of data extracted and transformed to be fed into a machine learning model, and contain information that is important for analysis and prediction.

[1163] A "machine learning model" is an algorithm or model that is trained using data to make predictions or classifications based on future data.

[1164] "Training means" refers to the means for training a machine learning model using collected data to improve the accuracy of the model.

[1165] A "means for predicting future trends" is a means for predicting trends in future data or events using a trained machine learning model.

[1166] A "user interface" is an interface through which a user accesses a system to input information or obtain information from the system.

[1167] The "means for converting into a displayable format" refers to a means for converting the prediction results into a visually easy-to-understand format (for example, a graph or chart) and displaying it on a user interface.

[1168] "Feedback" refers to predictions, opinions, and insights provided by users and experts.

[1169] "Means for obtaining feedback" refers to the means for collecting and storing feedback from users and experts.

[1170] "Retraining" refers to the process of retraining a machine learning model using feedback and new data to improve the model's accuracy.

[1171] "Platform" means a designated environment or system for experts to input and integrate their insights and analytical reports into the system.

[1172] "Insights" refers to detailed analysis and opinions provided by experts, and are important information for improving forecast accuracy.

[1173] The system of the present invention combines a multi-step process, including data collection, data analysis and trend prediction, receiving and processing user feedback, integrating expert insights, and presenting the prediction results. As a specific implementation method of the present invention, the following steps are described in detail.

[1174] Data collection

[1175] The server collects data in real time from multiple data sources, such as news feeds, social media, and economic indicators, via APIs. For example, the server uses Twitter's API to collect tweets based on specific keywords. It converts this data into a data frame using the Pandas library. The server then combines multiple data frames to create a unified dataset, which is then stored in an SQL database.

[1176] Data analysis and trend forecasting

[1177] The server reads the stored data from the database and extracts features. For example, the server extracts the number of tweets in the past 24 hours and the percentage of tweets with positive context. The server then trains a machine learning model such as random forest or LSTM using Scikit-learn or TensorFlow. The trained model is used to predict future trends. The prediction results are formatted into a visually easy-to-understand format (e.g., graphs or charts) using the Plotly library and sent to the user's device.

[1178] Receiving and processing feedback

[1179] Users use their own devices to input their predictions and opinions into the system. For example, a user can input a prediction such as, "I think sales will increase next quarter." The device then sends the input feedback to the server. The server stores the received feedback in a database and updates the training data based on the stored feedback. The server then retrains the machine learning model using the updated data.

[1180] Integrating expert insights

[1181] Experts access the platform and input their own insights and analysis reports. The device then sends these insights to the server, which stores them in a database and retrains the machine learning model based on the stored insights. This incorporates expert knowledge into the model, improving its prediction accuracy.

[1182] Prediction results

[1183] The user's device sends a request to the server to get the latest trend forecast information, and the server returns the latest forecast results, which the device displays on a dashboard in the form of interactive graphs and charts.

[1184] Specific examples

[1185] For example, a user can enter the following prompt into the system:

[1186] "Collect the latest tweet data and make supply and demand forecasts."

[1187] "Create a sales forecast for the next quarter and display it in a chart."

[1188] "Update the AI ​​model based on user feedback."

[1189] In response to these prompts, the server collects, analyzes, and processes appropriate data. Furthermore, by incorporating expert insights into the model, forecast accuracy can be improved. In this way, the system of the present invention helps companies respond quickly to market changes and make effective decisions.

[1190] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1191] Step 1: Data collection

[1192] The server uses APIs to collect data in real time from various data sources, such as news feeds, social media, and economic indicators. For example, the Twitter API is used to collect tweets based on specific keywords. The server periodically executes scheduled tasks to obtain the data. The collected data is converted into a data frame using the Pandas library. The input is raw data from each data source, and the output is a formatted data frame.

[1193] Step 2: Format and save the data

[1194] The server reformats the collected data using the Pandas library. This reformatting process includes removing duplicate data, handling missing values, and normalizing text. The reformatted data is then combined to create a unified dataset. This dataset is then stored in a SQL database. The input is an unformed data frame, and the output is the reformatted and combined dataset.

[1195] Step 3: Feature extraction

[1196] The server reads the stored data from the database and extracts features, such as the number of tweets per hour or the percentage of tweets with positive context. The Scikit-learn library is used to encode these features. The input is the dataset, and the output is a list of numerically encoded features.

[1197] Step 4: Train the model

[1198] The server trains the extracted features using machine learning models such as random forests and LSTM. For example, it uses past tweet data to predict future trends. Libraries such as TensorFlow and Scikit-learn are used for this training. The accuracy of the trained model is verified and saved. The input is a list of features, and the output is a trained machine learning model.

[1199] Step 5: Predicting trends

[1200] The server uses the trained model to predict future trends, for example, forecasting sales for the next quarter. The prediction results are visualized using the Plotly library and output in a format that is easy for users to understand. The input is the machine learning model and features, and the output is the visualized prediction results.

[1201] Step 6: Receiving feedback

[1202] Users input their predictions and opinions into the system through their terminals. For example, they input a prediction such as, "I think sales will increase next quarter." The terminals collect this feedback and send it to the server. The input is the user's feedback, and the output is the feedback data sent to the server.

[1203] Step 7: Process feedback and retrain

[1204] The server stores the received feedback in a database and updates the training data based on the stored feedback. The AI ​​model is retrained using the updated data, which improves the model's accuracy. The input is the feedback data, and the output is an updated machine learning model.

[1205] Step 8: Integrating expert insights

[1206] Experts access the platform and input their insights and analysis reports. The device sends them to the server, which stores the received insights in a database. The stored insights are used to retrain the machine learning model, further improving its prediction accuracy. The input is the expert's insights, and the output is the retrained machine learning model.

[1207] Step 9: Presenting the prediction results

[1208] The user's device requests the latest forecast results from the server. The server returns the latest forecast results, and the device displays these results on a dashboard as interactive graphs and charts, allowing the user to understand the latest market trends at a glance. The input is a request for forecast results, and the output is the forecast results in a displayable format.

[1209] (Application example 1)

[1210] 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."

[1211] Conventional ad delivery systems lacked real-time market trend analysis, making effective targeting and personalization difficult. They also struggled to quickly incorporate user and expert feedback, delaying the optimization of ad campaigns. Furthermore, limited means of collecting and analyzing user behavior data made it difficult to accurately evaluate advertising effectiveness.

[1212] 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.

[1213] In this invention, the server includes a data collection means, a means for formatting the collected data and storing it in a database, a means for extracting features from the stored data and training an AI model, a means for predicting future trends using the trained AI model, a means for sending the prediction results to a user interface and converting them into a displayable format, a means for receiving feedback from users and retraining the AI ​​model, a means for receiving insights and analysis reports from experts and storing them in a database and retraining the AI ​​model, and a means for generating personalized advertisements based on the collected and analyzed data and displaying them on user terminals. This enables effective targeting based on real-time market trend analysis, optimization of advertising campaigns that quickly reflect user and expert feedback, and accurate evaluation of advertising effectiveness by collecting and analyzing user behavior data.

[1214] "Data collection means" refers to a means for obtaining data in real time from multiple data sources such as news feeds, social media, and economic indicators.

[1215] "Database" means a centralized data repository for storing uniform data sets for later use in analysis and prediction.

[1216] A "feature extraction method" is a method for identifying information necessary for trend prediction from collected data and converting it into a format that can be used to train an AI model.

[1217] "AI model training method" refers to a method of training an AI model (such as random forest or LSTM) using extracted features to enable it to predict future trends.

[1218] A "trend forecasting tool" is a tool for forecasting future trends using a trained AI model.

[1219] The "user interface transmission means" is a means for converting the prediction results into a format that is easy for the user to understand and transmitting it to the user's terminal.

[1220] "Feedback acquisition means" refers to the means of collecting predictions and opinions from users and reflecting them in the system.

[1221] "AI model retraining method" refers to a method of retraining an AI model based on feedback from users and experts to improve its accuracy.

[1222] The "means for receiving expert insights" is a means for receiving insights and analysis reports from experts, storing them in a database, and then retraining the AI ​​model.

[1223] "Personalized advertisement generation means" refers to a means for generating advertisements optimized for individual users based on collected and analyzed data and displaying them on the user's terminal.

[1224] The system that realizes this application example comprises the smartphone app "AdTrend." The server collects data in real time from multiple data sources, such as news feeds, social media, and economic indicators. The collected data is formatted and converted into a data frame, and these data frames are then combined to create a unified dataset. This process uses Python's Pandas framework.

[1225] Next, the server reads the stored data from the database and extracts the necessary features. Based on the extracted features, it trains an AI model using algorithms such as random forest and LSTM. This training generates an AI model that can predict future trends. The model is trained using the Scikit-learn library and generates prediction results.

[1226] The server then formats these forecasts in JSON format and sends them to the user's smartphone, where they are displayed as interactive charts and graphs. This dashboard helps users visually understand trends.

[1227] Users can enter their predictions and opinions into the system through a form. This feedback is sent to the server and stored in a database. The stored feedback is used as new training data to retrain the AI ​​model. The Flask framework is used to receive and process user feedback.

[1228] In addition, insights and analysis reports from experts are also sent to the server and stored in the database, allowing the experts' knowledge to be utilized to improve the prediction accuracy of the AI ​​model. In this process, the experts' insights are incorporated into the retraining of the AI ​​model.

[1229] Regarding the generation of personalized ads, the collected data is used to generate ads optimized for individual user attributes. These ads are delivered and displayed on the user's device. User behavior data is collected and analyzed to evaluate the effectiveness of the ads. Ad campaigns are optimized using Python and related libraries.

[1230] Examples of specific prompts include:

[1231] Trend Forecast:

[1232] Analyze the characteristics of the most popular advertising campaigns in the next quarter and propose effective advertising strategies.

[1233] Feedback received:

[1234] We collect user opinions about the ads currently being served and use that data to personalize ads.

[1235] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1236] Step 1:

[1237] The server collects data in real time from multiple data sources, such as news feeds, social media, and economic indicators. It uses an API to retrieve the data and converts it into a Pandas data frame. The input is raw data from each data source, and the output is organized data in a data frame format.

[1238] Step 2:

[1239] The server combines the multiple formatted data frames to create a unified dataset and stores it in the database. The input is multiple data frames, and the output is the combined unified dataset. Specifically, the server combines the data frames using functions such as the merge function in Pandas and stores it in the database.

[1240] Step 3:

[1241] The server reads the stored data from the database and extracts the necessary features. The input is a unified dataset, and the output is the features required for training the AI ​​model. Specifically, it uses Pandas and Scikit-learn to extract the features and prepare them as training data.

[1242] Step 4:

[1243] The server uses the extracted features to train an AI model. The input is the features, and the output is a trained AI model (e.g., random forest or LSTM). Specifically, the server uses the Scikit-learn library to train the model and check its prediction accuracy.

[1244] Step 5:

[1245] The server uses a trained AI model to predict future trends. The input is newly collected data, and the output is the future trend prediction result. Specifically, new data is input into the model and a prediction result is generated.

[1246] Step 6:

[1247] The server formats the prediction results in JSON format and sends them to the user's device. The input is the prediction result data, and the output is the JSON-formatted prediction result sent to the user's device. Specifically, Flask is used to communicate data between the server and the device.

[1248] Step 7:

[1249] The terminal displays the received prediction results as interactive charts and graphs. The input is the prediction result data in JSON format, and the output is the prediction result displayed graphically. Specifically, a dashboard is created using a JavaScript library (e.g., D3.js) to provide visual feedback to the user.

[1250] Step 8:

[1251] Users enter their predictions and opinions into the system through a form. The input is the user's feedback, and the output is the feedback data sent to the server. Specifically, the system uses a combination of HTML forms and JavaScript to send user input to the server.

[1252] Step 9:

[1253] The server receives feedback from the user and stores it in a database. The input is the feedback data, and the output is the feedback stored in the database. Specifically, Flask is used to receive the data and perform the storage process.

[1254] Step 10:

[1255] The server then retrains the AI ​​model using the stored feedback. The input is the feedback data in the database, and the output is the retrained AI model. Specifically, it repeats feature extraction and model training as before.

[1256] Step 11:

[1257] The expert inputs his / her insights and analysis reports and sends them to the server. The input is the expert's insights and analysis reports, and the output is the insight data sent to the server. Specifically, the system accepts the expert's input via a form.

[1258] Step 12:

[1259] The server stores the insights received from the experts in a database and uses them to retrain the AI ​​model. The input is the expert's insight data, and the output is a retrained AI model that reflects the insights. Specifically, the insight data is stored in a database and used for the re-learning process.

[1260] Step 13:

[1261] The server generates personalized advertisements based on the user's attribute data and displays them on the user's device. The input is the user's attribute data, and the output is personalized advertisements. Specifically, the advertisement content is dynamically changed based on the attribute data and delivered to the user.

[1262] Step 14:

[1263] The server collects user behavior data and evaluates the effectiveness of advertising. The input is user behavior data, and the output is evaluated advertising effectiveness data. Specific operations include analyzing the behavior data and calculating evaluation indicators.

[1264] 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.

[1265] The system for implementing the present invention consists of a multi-step process that combines data collection, data analysis and trend prediction, receiving and processing user feedback, integrating expert insights, presenting prediction results, and an emotion engine that recognizes user emotions.

[1266] Data collection and formatting

[1267] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. It formats this data and converts it into a data frame format. It combines multiple data frames to create a unified dataset and stores this dataset in a database. As an example, it retrieves the latest business news from a news API and converts it into a data frame.

[1268] Feature extraction and AI model training

[1269] The server reads the stored data from the database and extracts the necessary features. An AI model is trained based on the extracted features. This AI model uses algorithms such as random forest and LSTM. Features include market prices, sales data, and seasonal factors.

[1270] Future trend predictions

[1271] Using the trained AI model, the server predicts future trends. The prediction results are displayed in a visually easy-to-understand format (for example, graphs or charts). This allows users to understand the prediction results at a glance. For example, the sales forecast for the next quarter is displayed in chart format.

[1272] Prediction results

[1273] When a user accesses the device, it sends a request to the server to get the latest trend forecast information. The server returns the forecast results to the device, which then displays them on a dashboard in the form of interactive graphs and charts. For example, the UI displays real-time stock price forecasts in infographic format.

[1274] Receiving and processing user feedback

[1275] Users access the system through their own devices and input their predictions and opinions. This feedback is sent to the server and stored in a database. The stored feedback is then used to retrain the AI ​​model. For example, a user's opinion such as "the next sales will increase" is incorporated as retraining data.

[1276] Integrating expert insights

[1277] Experts access the platform and input their own insights and analysis reports, which are then sent to the server and stored in a database.The AI ​​model is then retrained based on the stored insights to improve prediction accuracy.

[1278] Collecting and processing user sentiment data

[1279] The server collects user emotional data through the emotion engine. For example, when a user enters feedback, the server analyzes emotions from their facial expressions and text to obtain emotional data. This emotional data is then stored in a database along with the user's feedback and opinions.

[1280] Retraining using emotion data

[1281] The server retrains the AI ​​model based on the emotional data obtained from the emotion engine. This retraining process makes it possible to provide prediction results that take the user's emotional state into account, further improving prediction accuracy.

[1282] Specific examples

[1283] For example, the server retrieves business news from a news API, converts it into a data frame, and combines it. Features are extracted and used to train an AI model. Future trends are then predicted and displayed in graph form. When a user accesses the system on their device, this predicted data is displayed on a dashboard. Users input their own opinions and sentiment data, which is sent to the server and used for retraining. Experts also provide insights, improving the overall accuracy of the AI ​​model.

[1284] This is the specific processing flow of this system. By adopting this system, companies will be able to grasp market trends more precisely and in real time, enabling them to make effective decisions.

[1285] The processing flow will be explained below.

[1286] Step 1:

[1287] Data collection and formatting (server)

[1288] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. The retrieved data is converted from JSON format to a data frame format. The server then combines multiple data frames to create a unified dataset and stores this dataset in a database. For example, data retrieved from an economic indicator API is updated daily and stored in a database.

[1289] Step 2:

[1290] Feature extraction (server)

[1291] The server reads the stored data from the database and extracts the necessary features. These features range from sales data, consumer sentiment trends, seasonal factors, product ratings, etc. These features are necessary for the AI ​​model to accurately predict trends.

[1292] Step 3:

[1293] AI model training (server)

[1294] The server trains an AI model based on the extracted features. This AI model uses algorithms such as random forest and LSTM. It uses daily sales data and economic indicators as features and learns the importance of each variable.

[1295] Step 4:

[1296] Future Trend Prediction (Server)

[1297] The server uses the trained AI model to predict future trends. The prediction results are displayed in a visually easy-to-understand format (e.g., graphs and charts), allowing users to understand the prediction results at a glance.

[1298] Step 5:

[1299] Presenting prediction results (device)

[1300] When a user accesses the device, it sends a request to the server to get the latest trend forecast information. The server then sends the forecast results to the device, which then displays them on a dashboard in the form of interactive graphs and charts. For example, the next quarter's sales forecast can be displayed in infographic format.

[1301] Step 6:

[1302] Receiving user feedback (user, device)

[1303] Users access the system through their own terminals and input their predictions and opinions. For example, they can input a prediction such as, "I think sales will increase next quarter." This feedback is sent to the server through a form.

[1304] Step 7:

[1305] Feedback storage and processing (server)

[1306] The server receives the feedback sent by the user and stores it in a database. It then retrains the AI ​​model based on the stored feedback. This retraining process improves the accuracy of the prediction model.

[1307] Step 8:

[1308] Integration of expert insights (server, user)

[1309] Experts access the platform and input their own insights and analytical reports, which the server then stores in a database and uses to retrain the AI ​​model, incorporating the experts' knowledge to further improve its prediction accuracy.

[1310] Step 9:

[1311] Collecting user emotion data (device, user)

[1312] When the user inputs feedback, the device collects the user's emotional data through the emotion engine. For example, when the user inputs feedback, the device analyzes emotions from the user's facial expressions and text to obtain the emotional data.

[1313] Step 10:

[1314] Storage and processing of emotional data (server)

[1315] The server receives the emotion data sent from the device and stores it in a database along with the feedback data. The emotion data is then used to retrain the AI ​​model. This process allows it to provide prediction results that take the user's emotional state into account.

[1316] This is the specific processing flow of this system, which enables companies to make accurate predictions based on multifaceted data, including user sentiment, and supports more effective and faster decision-making.

[1317] Example 2

[1318] 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."

[1319] In today's business environment, accurately predicting market trends is extremely difficult, and companies require real-time analysis and forecasting to make effective decisions. Traditional forecasting systems struggle to take user feedback and sentiment into account, resulting in poor forecast accuracy. Furthermore, the lack of a means to effectively integrate expert insights limits overall forecast accuracy.

[1320] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a data collection means, a means for formatting the collected data and storing it in an information storage device, a means for extracting features based on the stored data and training a generative learning model, a means for predicting future trends using the trained generative learning model, a means for transmitting the prediction results to a terminal device interface and converting them into a displayable format, a means for obtaining feedback from users and retraining the generative learning model, a means for collecting user emotion data using an emotion recognition device, and a means for retraining the generative learning model based on the collected emotion data. This makes it possible to collect and format information from multiple data sources in real time and accurately predict future trends using the generative learning model. Furthermore, taking user feedback and emotions into account improves prediction accuracy, and integrating it with expert insights can further improve prediction accuracy.

[1321] "Data collection means" refers to a means for obtaining data in real time from multiple data sources such as news feeds, social media, and economic indicators.

[1322] An "information storage device" is a database or storage system for saving and managing formatted data.

[1323] "Features" are key data elements extracted from stored data and used to train predictive models.

[1324] A "generative learning model" is an AI model that learns from training data and predicts future trends.

[1325] "Endpoint interface" refers to the interactive display screen or dashboard that a user uses to view prediction results.

[1326] "User feedback" refers to predictions and opinions entered by users through the system.

[1327] "Retraining" is the process of adding new data to an existing training model to improve the model's predictive accuracy.

[1328] An "emotion recognition device" is an engine or system that analyzes and acquires emotional data from user feedback, facial expressions, and text.

[1329] "Emotional data" refers to data that represents a user's emotional state, including emotional categories such as positive, negative, and neutral.

[1330] "Trend forecasting" refers to the process of using trained generative learning models to predict future market movements and trends.

[1331] The present invention is a system for forecasting future market trends by collecting data from multiple data sources in real time, training and relearning a generative learning model based on the collected data, and integrating feedback from servers, terminals, and users to improve forecast accuracy.

[1332] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. It uses common data processing libraries (e.g., Pandas, NumPy) to format this data and convert it into a data frame. It then combines these data frames to create a unified dataset and stores it in a database (e.g., MySQL, PostgreSQL).

[1333] Next, the server reads the stored data from the database and extracts the necessary features. This extraction uses a data analysis library (e.g., Scikit-learn, TensorFlow). Based on the extracted features, a generative learning model (e.g., Random Forest, LSTM) is trained. This training process allows the AI ​​model to learn trends from past data and predict future trends.

[1334] Using the trained generative learning model, the server predicts future trends and formats the prediction results in a visually understandable format such as graphs and charts. This visualization is done using visualization libraries (e.g., Matplotlib, Plotly). The prediction results are sent to the end device interface and displayed on an interactive dashboard when accessed by the user.

[1335] Users access the system through their devices and input their predictions and opinions. This feedback is sent to the server and stored in a database. Based on the stored feedback, the generative learning model is retrained to improve prediction accuracy. This retraining process uses techniques to update the existing model using new data.

[1336] The system also includes an emotion recognizer that collects emotion data from user feedback, facial expressions, and text. Emotion recognition uses natural language processing libraries (e.g., NLTK, spaCy) and face recognition libraries (e.g., OpenCV). The collected emotion data is stored in a database and used to retrain a generative learning model. This allows prediction results to be provided that take the user's emotional state into account.

[1337] As a concrete example, the server retrieves the latest business news from a news API, converts the article contents into a data frame, and combines them. It then extracts features from market prices and performance data to train a generative learning model. It predicts sales for the next quarter and displays the results in a graph format. When accessed by a user, the predicted data is displayed on the device's dashboard, and the user can enter their own feedback and sentiment data. This is sent to the server and used for re-training. Experts also provide insights, improving the accuracy of the generative learning model.

[1338] Prompt Sentence Examples

[1339] "Please use the Random Forest algorithm to predict sales for the next quarter and display the results in a graph."

[1340] "Predict market trends based on sentiment data and present the results to users in infographics."

[1341] "Retrain your generative learning model with user feedback to improve prediction accuracy."

[1342] By adopting this system, companies will be able to collect information from multiple data sources in real time, integrate the opinions of users and experts, and gain a more precise understanding of market trends in real time, enabling them to make more effective decisions.

[1343] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1344] System program processing flow

[1345] Step 1: Data collection

[1346] The server retrieves data in real time from multiple data sources, such as news feeds, social media, and economic indicators. Specifically, it sends API requests to retrieve news articles, social media posts, and economic indicator data. As input, it uses API endpoints and access keys. As output, it returns raw data to the server.

[1347] Step 2: Data Formatting and Storage

[1348] The server formats the retrieved data and converts it into a data frame. Specifically, it extracts information such as the title, content, and date and time of the retrieved news articles and formats them into a data frame using the Pandas or NumPy library. Raw data is used as input, and a data frame is obtained as output. This is then saved in a database. For example, it connects to a database such as MySQL or PostgreSQL and inserts the data.

[1349] Step 3: Feature extraction

[1350] The server reads the stored data from the database and extracts the necessary features. Specifically, features such as market prices, sales data, and seasonal factors are calculated using scripts and extracted using data analysis libraries (e.g., Scikit-learn, TensorFlow). The data read from the database is used as input, and the features are obtained as output.

[1351] Step 4: Training the AI ​​model

[1352] The server trains a generative AI model based on the extracted features. Algorithms used include random forest and LSTM. Specifically, the model is trained using a training dataset and the trained model is saved. The feature dataset is used as input, and the trained model is obtained as output.

[1353] Step 5: Trend forecasting

[1354] The server uses a trained generative AI model to predict future trends. Specifically, it inputs new data into the trained model to predict future market trends and sales. New data is used as input, and predictions are obtained as output. The predictions are then formatted in a visually easy-to-understand format (e.g., graphs or charts).

[1355] Step 6: Formatting and presenting prediction results

[1356] When a user accesses the device, it sends a request to obtain the latest trend forecast information from the server. The server returns the forecast results to the device, which then displays them on a dashboard in the form of interactive graphs and charts. For example, the forecast results can be visualized using libraries such as Plotly or D3.js. The forecast results are used as input, and a visualized dashboard is obtained as output.

[1357] Step 7: Collect and process user feedback

[1358] Users access the system through their own devices and enter their predictions and opinions. A feedback form is provided as input, and users enter their opinions and predictions. This feedback is sent to the server and stored in a database. The saved feedback data is obtained as output.

[1359] Step 8: Integrating expert insights

[1360] Experts access the platform and input their own insights and analytical reports. These insights are sent to the server and stored in the database. The expert insights are used as input and the stored insight data is obtained as output. The generative AI model is then retrained based on the stored insights to improve prediction accuracy.

[1361] Step 9: Collect and process emotion data

[1362] The server uses an emotion recognition device to collect user emotion data. Specifically, when the user enters feedback, it analyzes emotions from their facial expressions and text. It uses libraries such as OpenCV and NLTK. It uses the user's feedback data and facial expression data as input, and obtains analyzed emotion data as output. This is then stored in a database.

[1363] Step 10: Retraining with emotion data

[1364] The server retrains the generative AI model based on the collected emotional data. Specifically, the emotional data is incorporated into the existing learning model as features to improve prediction accuracy. The emotional data and feature data are used as input, and a retrained generative AI model is obtained as output. This provides prediction results that take the user's emotional state into account.

[1365] (Application example 2)

[1366] 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."

[1367] Conventional trend prediction systems have difficulty taking user feedback into account, particularly the influence of user emotions. Furthermore, there is a lack of a means to efficiently integrate real-time user feedback and expert insights, limiting the accuracy of predictions. This makes it difficult to provide product recommendations that meet user needs, especially on online shopping sites.

[1368] 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.

[1369] In this invention, the server includes a data collection means, a means for formatting the collected data and storing it in a database, a means for extracting features from the stored data and training an artificial intelligence model, a means for predicting future trends using the trained artificial intelligence model, a means for sending the prediction results to a user interface and converting them into a displayable format, a means for obtaining feedback from users and retraining the artificial intelligence model, a means for collecting user emotional data and storing it in a database together with the feedback, and a means for retraining the artificial intelligence model based on the emotional data. This enables more accurate trend predictions that take the user's emotional state into account and personalized product recommendations for each user.

[1370] "Data collection means" refers to the ability to obtain information in real time from external data sources.

[1371] "Data formatting means" refers to the function of converting collected raw data into a usable format and storing it in a database.

[1372] "Feature extraction means" refers to the function that extracts important data points from stored data and uses them to train artificial intelligence models.

[1373] An "artificial intelligence model" refers to an algorithm or system that has been trained to make predictions or analyses.

[1374] "Trend Forecasting" refers to the ability to predict future trends using a trained artificial intelligence model.

[1375] "User interface" refers to a function or screen that visually presents prediction results to the user and allows them to operate them interactively.

[1376] "Feedback acquisition means" refers to the function of collecting opinions and advice from users and incorporating them into the system.

[1377] "Re-learning means" refers to the function of re-training the AI ​​model based on collected feedback to improve prediction accuracy.

[1378] "Emotional data" refers to information about a user's emotional state analyzed from their text and images.

[1379] "Emotional data collection means" refers to the function of analyzing the user's emotional state and storing it in a database.

[1380] The system for implementing the present invention consists of the following steps: data collection, data analysis and trend prediction, receiving and processing user feedback, collecting and integrating emotion data, and re-training. Each step is described in detail below.

[1381] Data collection

[1382] The server sends API requests to retrieve data in real time from multiple data sources, such as news feeds, social media, and economic indicators. For example, news APIs, social media APIs, and economic indicator APIs are used. These raw data are converted into a data frame format and combined to create a unified dataset. This dataset is then stored in a database system, such as MySQL or PostgreSQL.

[1383] Feature extraction and AI model training

[1384] The server reads the stored data from the database and extracts the necessary features. Technically, this can be done using Python data analysis libraries such as pandas or numpy. Based on the extracted features, an artificial intelligence model is trained using a machine learning library such as scikit-learn. This AI model can use, for example, random forest or LSTM.

[1385] Future trend predictions

[1386] Using trained artificial intelligence models, the server predicts future trends. The results are then formatted into easy-to-understand visual displays (e.g., graphs and charts). For example, sales forecasts for the next quarter are displayed in chart format.

[1387] Prediction results

[1388] When a user accesses the device, it sends a request to the server to get the latest trend forecast information. The server returns the forecast results to the device, which then displays them on a dashboard in the form of interactive graphs and charts. For example, the UI displays real-time stock price forecasts in infographic format.

[1389] Receiving and processing user feedback

[1390] Users access the system through their devices and input their predictions and opinions. This feedback is entered, for example, using an HTML form and sent to the server. The server receives it and stores it in a database. The stored feedback is then used to retrain the AI ​​model.

[1391] Collecting and processing emotional data

[1392] The server collects user emotion data through an emotion recognition engine. For example, libraries such as TextBlob and facial_emotion_recognition are used to analyze the emotion of facial expressions and text when users enter feedback. The resulting emotion data is then stored in a database along with the feedback.

[1393] Retraining using emotion data

[1394] The server retrains the AI ​​model based on the emotion data obtained from the emotion engine. This retraining process allows the model to provide predictions that take the user's emotional state into account, further improving prediction accuracy.

[1395] As a concrete example, the server retrieves business news from a news API, converts it into a data frame, and combines it. Features are extracted and an AI model is trained. Next, future trends are predicted and displayed in graph form. When a user accesses the system on their device, this predicted data is displayed on a dashboard. The user enters emotional data along with their own opinion, which is sent to the server and used for re-learning. An example of a prompt used here is, "Do you think sales will increase in the next quarter? Please explain why. Please also upload an image of your feedback."

[1396] The above is a specific embodiment of the present system. By adopting this system, companies can grasp market trends more precisely and in real time, enabling them to make effective decisions.

[1397] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1398] Step 1:

[1399] The server sends API requests from multiple data sources, such as news feeds, social media, and economic indicators, and retrieves data in real time.

[1400] Input: API request

[1401] Data processing: Convert the response from each API into a data frame format

[1402] Output: Data in data frame format

[1403] Step 2:

[1404] The server formats the acquired data into multiple data frames and combines them to create a unified dataset.

[1405] Input: Dataframe of news, social media, and economic indicator data

[1406] Data processing: Combining data frames

[1407] Output: Unified dataset

[1408] Step 3:

[1409] The server stores the unified data set in a database.

[1410] Input: Unified dataset

[1411] Action: Insert data into database

[1412] Output: Saved

[1413] Step 4:

[1414] The server reads the stored data from the database and extracts the necessary features.

[1415] Input: Data read from database

[1416] Data processing: feature extraction process

[1417] Output: Extracted features

[1418] Step 5:

[1419] The server trains an artificial intelligence model based on the extracted features.

[1420] Input: extracted features

[1421] Data processing: training artificial intelligence models

[1422] Output: A trained artificial intelligence model

[1423] Step 6:

[1424] The server uses trained artificial intelligence models to predict future trends.

[1425] Input: A trained AI model and the data to be predicted

[1426] Data calculation: Future trend prediction process

[1427] Output: Prediction results

[1428] Step 7:

[1429] The server formats the prediction results into interactive graphs and charts and sends them to the user interface.

[1430] Input: Prediction result

[1431] Data processing: Conversion into infographics format

[1432] Output: Interactive graphs and charts

[1433] Step 8:

[1434] The device displays the prediction results in the user interface and is ready to receive feedback from the user.

[1435] Input: Interactive graphs and charts

[1436] Behavior: Show in user interface

[1437] Output: Displayed prediction results

[1438] Step 9:

[1439] Users enter their predictions and opinions through a form and also upload image feedback.

[1440] Input: User opinions and feedback images

[1441] Action: Enter and send feedback

[1442] Output: Feedback data

[1443] Step 10:

[1444] The server performs text analysis on the received feedback to obtain emotion data.

[1445] Input: User feedback

[1446] Data Computing: Text Analysis and Emotion Recognition

[1447] Output: Emotion data

[1448] Step 11:

[1449] The server stores the acquired emotion data and feedback in a database.

[1450] Input: Emotion data and feedback

[1451] Action: Insert data into database

[1452] Output: Saved

[1453] Step 12:

[1454] The server retrains the artificial intelligence model based on the stored feedback and emotion data.

[1455] Input: Feedback and emotion data

[1456] Data Computing: Retraining Artificial Intelligence Models

[1457] Output: Updated artificial intelligence model

[1458] The above is the specific operation and data flow of each processing step in this system.

[1459] 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.

[1460] 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.

[1461] 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.

[1462] 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.

[1463] 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.

[1464] 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.

[1465] 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).

[1466] 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.

[1467] 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."

[1468] 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.

[1469] 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).

[1470] 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.

[1471] 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.

[1472] 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.

[1473] 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.

[1474] 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.

[1475] 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.

[1476] 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.

[1477] 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.

[1478] 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.

[1479] 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.

[1480] The following is further disclosed regarding the above embodiment.

[1481] (Claim 1)

[1482] data collection means;

[1483] A means of formatting the collected data and storing it in a database;

[1484] A means to extract features from the stored data and train an AI model,

[1485] A means of predicting future trends using trained AI models;

[1486] means for transmitting the prediction results to a user interface and converting them into a displayable format;

[1487] and a means for obtaining user feedback and retraining the AI ​​model.

[1488] (Claim 2)

[1489] A means of obtaining data from data sources in real time; and

[1490] A means for converting and combining the acquired data into multiple data frames;

[1491] and means for storing the combined data in a database.

[1492] (Claim 3)

[1493] A means of receiving user-entered predictions and opinions through a form;

[1494] a means for storing the received predictions and opinions in a database;

[1495] and means for retraining the AI ​​model using the stored feedback.

[1496] "Example 1"

[1497] (Claim 1)

[1498] data collection means;

[1499] A means of formatting the collected data and storing it in a database;

[1500] A means to extract features from the stored data and train machine learning models;

[1501] A means of predicting future trends using trained machine learning models; and

[1502] means for transmitting the prediction results to a user interface and converting them into a displayable format;

[1503] A means to obtain user feedback and retrain the machine learning model;

[1504] a means of receiving expert insights through the platform and storing them in a database;

[1505] and means for retraining the machine learning model using the stored expert insights to improve predictive accuracy.

[1506] (Claim 2)

[1507] A means of obtaining data from data sources in real time; and

[1508] A means for converting and combining the acquired data into multiple data frames;

[1509] and means for storing the combined data in a database.

[1510] (Claim 3)

[1511] A means of receiving user-entered predictions and opinions through a form;

[1512] a means for storing the received predictions and opinions in a database;

[1513] and means for retraining the machine learning model using the stored feedback.

[1514] "Application Example 1"

[1515] (Claim 1)

[1516] data collection means;

[1517] A means of formatting the collected data and storing it in a database;

[1518] A means to extract features from the stored data and train an AI model,

[1519] A means of predicting future trends using trained AI models;

[1520] means for transmitting the prediction results to a user interface and converting them into a displayable format;

[1521] A means to obtain user feedback and retrain the AI ​​model;

[1522] A means to receive insights and analytical reports from experts, store them in a database, and retrain the AI ​​model.

[1523] and means for generating personalized advertisements based on the collected and analyzed data and displaying them on the user terminal.

[1524] (Claim 2)

[1525] A means of obtaining data from data sources in real time; and

[1526] A means for converting and combining the acquired data into multiple data frames;

[1527] and means for storing the combined data in a database.

[1528] (Claim 3)

[1529] A means of receiving user-entered predictions and opinions through a form;

[1530] a means for storing the received predictions and opinions in a database;

[1531] A means to retrain the AI ​​model using the stored feedback; and

[1532] Collecting and analyzing user behavior data to assess the effectiveness of personalized advertising;

[1533] and means for incorporating expert insights to optimize the advertising campaign.

[1534] "Example 2: Combining Emotion Engines"

[1535] (Claim 1)

[1536] data collection means;

[1537] A means for formatting the collected data and storing it in an information storage device;

[1538] A means for extracting features from the stored data and training a generative learning model;

[1539] a means of predicting future trends using the trained generative learning model;

[1540] means for transmitting the prediction results to a terminal interface and converting them into a displayable format;

[1541] a means for obtaining user feedback and retraining the generative learning model;

[1542] means for collecting emotion data of a user using an emotion recognition device;

[1543] and means for retraining the generative learning model based on the collected emotion data.

[1544] (Claim 2)

[1545] a means of obtaining data from sources in real time;

[1546] A means for converting and combining the acquired data into multiple data frames;

[1547] and means for storing the combined data in an information storage device.

[1548] (Claim 3)

[1549] A means for receiving predictions and opinions entered by users through a form;

[1550] means for storing the received predictions and opinions in an information storage device;

[1551] and means for retraining the generative learning model using the stored feedback.

[1552] "Application example 2 when combining emotion engines"

[1553] (Claim 1)

[1554] data collection means;

[1555] A means of formatting the collected data and storing it in a database;

[1556] A means of extracting features from the stored data and training an artificial intelligence model;

[1557] A means of predicting future trends using trained artificial intelligence models; and

[1558] means for transmitting the prediction results to a user interface and converting them into a displayable format;

[1559] A means to obtain user feedback and retrain the AI ​​model;

[1560] A means of collecting user sentiment data and storing it in a database along with their feedback;

[1561] and means for retraining the artificial intelligence model based on the emotion data.

[1562] (Claim 2)

[1563] A means of obtaining data from data sources in real time; and

[1564] A means for converting and combining the acquired data into multiple data frames;

[1565] and means for storing the combined data in a database.

[1566] (Claim 3)

[1567] A means of receiving user-entered predictions and opinions through a form;

[1568] a means for storing the received predictions and opinions in a database;

[1569] and means for retraining the artificial intelligence model using the stored feedback. [Explanation of symbols]

[1570] 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. data collection means; A means of formatting the collected data and storing it in a database; A means to extract features from the stored data and train an AI model, A means of predicting future trends using trained AI models; means for transmitting the prediction results to a user interface and converting them into a displayable format; and a means for obtaining user feedback and retraining the AI ​​model.

2. A means of obtaining data from data sources in real time; and A means for converting and combining the acquired data into multiple data frames; and means for storing the combined data in a database.

3. A means of receiving user-entered predictions and opinions through a form; a means for storing the received predictions and opinions in a database; and means for retraining the AI ​​model using the stored feedback.

Images