system

Abstract

To provide a system for achieving more precise customer targeting by a means for quickly and accurately grasping a market trend, and for achieving advanced strategy planning based on the market trend, and for analyzing the interest or purchase action of a target audience in detail.SOLUTION: A means for selecting a data source from a user and accessing an external API or an internal database, a means for pre-processing collected data and filtering unnecessary information, a means for inputting the pre-processed data into a generative AI model to analyze a brand name, reputation, and attributes of a target audience, a means for displaying the analysis result on a dashboard in real time and providing it to a user, a means for predicting future trends and providing a prediction result, a means for generating marketing and business strategies based on the analysis result and the prediction result and notifying the user, and a means for obtaining feedback from the user and retraining the 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] Traditional market research and online surveys have been difficult to obtain accurate insights quickly due to limited information, low reliability, and the time and cost required for data collection and analysis. As a result, it has been difficult to develop appropriate marketing and sales strategies, making it difficult to target customers and implement campaigns effectively. There have also been challenges in predicting future market trends and building forward-looking strategies. [Means for solving the problem]

[0005] The system solves this problem by including a means for selecting data sources from users and accessing external APIs and internal databases, a means for preprocessing the collected data and filtering unnecessary information, a means for inputting the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target audience attributes, a means for displaying the analysis results on a dashboard in real time and providing them to users, a means for predicting future trends and providing the prediction results, a means for generating marketing and sales strategies based on the analysis and prediction results and notifying users, and a means for obtaining user feedback and retraining the model. This system enables rapid and accurate understanding of market trends and the development of advanced strategies based on them. Furthermore, detailed analysis of the target audience's hobbies and purchasing behavior enables more precise customer targeting.

[0006] A "data source" is a source of data that is the source of information that can be obtained from an external API, an internal database, etc.

[0007] An "external API" is an application program interface provided by another company or external organization, through which data can be obtained and manipulated.

[0008] An "internal database" is a system for managing and storing data held internally by a company or organization.

[0009] "Preprocessing" is the process of filtering unnecessary information from collected raw data and preparing it in a format that is easy to analyze.

[0010] A "generative AI model" is an artificial intelligence model that uses collected data to analyze brand awareness, reputation, and target audience attributes.

[0011] "Brand awareness" is an indicator of how well a particular brand is recognized in the market and among consumers.

[0012] "Reputation" refers to the positive or negative opinions and evaluations that the market and consumers have of a particular brand.

[0013] "Target audience attributes" are characteristics such as age, gender, hobbies, and purchasing behavior of a group of customers for a particular brand or product.

[0014] A "dashboard" is an interface that visualizes analysis results and real-time data, allowing users to intuitively understand the situation.

[0015] "Trend forecasting" is the use of generative AI models to predict future market trends and changes in brand awareness and reputation.

[0016] A "marketing and sales strategy" is a policy and plan for specific market and sales activities to be implemented to improve brand awareness and customer satisfaction.

[0017] "Feedback" refers to the evaluations and reactions received as a result of the marketing and sales strategies implemented by users, and is data used to improve the system and retrain the model.

[0018] "Retraining" is the process of retraining an AI model based on new data and feedback collected to improve its accuracy. [Brief explanation of the drawings]

[0019] [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

[0020] 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.

[0021] First, the terms used in the following description will be explained.

[0022] 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).

[0023] 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.

[0024] 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.

[0025] 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.

[0026] 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."

[0027] [First embodiment]

[0028] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0029] 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.

[0030] 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).

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0036] 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.

[0037] 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.

[0038] 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.

[0039] 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."

[0040] This invention is a system that utilizes generative AI to understand brand awareness, reputation, and target audience insights in real time to enhance marketing and sales strategies. This system is composed of multiple servers, terminals, and users.

[0041] System Configuration

[0042] 1. Data Collection Methods

[0043] The server accesses external APIs and internal databases to collect data related to a specific brand. For example, the server uses the Twitter API to collect tweets containing the brand name and related hashtags. At the same time, it retrieves past purchase history and customer information from the internal database.

[0044] 2. Data preprocessing methods

[0045] The server preprocesses the collected data and filters out unnecessary information. Specifically, the server removes spam messages and irrelevant data and standardizes the format of the text data. During this process, the server analyzes the text data using natural language processing technology.

[0046] 3. Data Analysis Methods

[0047] The server inputs the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target audience attributes. At this stage, the server uses a machine learning model to perform sentiment analysis and classify the data as positive, negative, or neutral. It also generates a target audience profile based on user purchasing behavior and hobbies.

[0048] 4. Real-time feedback methods

[0049] The server displays the analysis results in real time on a dashboard and provides them to the user. For example, the server uses WebSocket technology to continuously update the latest analysis results to the dashboard UI component, allowing users to grasp market trends in real time.

[0050] 5. Means of predicting future trends

[0051] Sarver uses generative AI models to predict future trends for brands. Based on time-series data, it predicts fluctuations in awareness and reputation over the next few months and displays the results on a dashboard. For example, Sarver predicts that brand awareness will increase by 5% next month.

[0052] 6. Strategy Generation Method

[0053] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user. Specifically, the server makes strategy suggestions such as "recommending providing a promotion code in the next campaign." These suggestions are notified to the user not only via the dashboard but also by email and push notifications.

[0054] 7. Feedback channels

[0055] Users provide feedback on the results of their campaigns and strategies to the server. The server uses this feedback to retrain the AI ​​model and improve the accuracy of the system. For example, a user can input the results of a campaign, and the server can collect that data and use it for the next model training.

[0056] Specific examples

[0057] For example, a beverage manufacturer might use this system when launching a new product. The server collects tweets related to the "new beverage name" from social media and analyzes its popularity in real time. The server also combines this with past purchasing data to identify that "men in their 30s who love sports" are the primary target audience. The generative AI model then predicts that the product's popularity will increase next month. Based on this information, the system suggests a strategy to the user, such as "running a promotion during a sporting event," and provides feedback on the results to be used next time.

[0058] The above is a specific embodiment of the present invention. By using this system, companies can quickly and accurately grasp market trends and implement effective marketing and sales strategies.

[0059] The processing flow will be explained below.

[0060] Step 1: Select a data source

[0061] A data source is selected so that the server can collect data from an external API (e.g., Twitter API) or an internal database. The server obtains an API key and authentication information to establish access to the external API.

[0062] Step 2: Getting the data

[0063] The server sends a request containing specific keywords or hashtags to retrieve data from selected data sources. Specifically, the server sends an API request to search tweets and posts for the brand name and related keywords.

[0064] Step 3: Preprocessing the data

[0065] The server preprocesses the retrieved data and filters out unnecessary information, specifically filtering out spam messages and normalizing the text (e.g., converting to lowercase and removing extra whitespace). In this step, natural language processing techniques are used to analyze the text data.

[0066] Step 4: Save your data

[0067] The server stores the pre-processed data in an internal database for quick access in subsequent processing steps.

[0068] Step 5: Perform sentiment analysis

[0069] The server uses a generative AI model to perform sentiment analysis on the pre-processed data, for example, taking tweet or review text as input and classifying it as positive, negative, or neutral.

[0070] Step 6: Target Audience Analysis

[0071] The server combines the results of the sentiment analysis with the user's purchasing history to perform a detailed analysis of the target audience's attributes, specifically creating a profile of their age group, gender, hobbies, and preferences.

[0072] Step 7: Provide real-time feedback

[0073] The server displays the analysis results in a dashboard in real time, using WebSocket or API to ensure that data is reflected to users in real time.

[0074] Step 8: Perform future trend predictions

[0075] The server uses a generative AI model to predict future trends for the brand, predicting fluctuations in name recognition and reputation based on time-series data.

[0076] Step 9: Providing prediction results

[0077] The server displays the prediction results on a dashboard and provides them to the user, visualizing them in graphs and charts for easy understanding.

[0078] Step 10: Generate a marketing and sales strategy

[0079] Based on the analysis and prediction results, the server proposes specific marketing and sales strategies to the user, such as providing promotion codes and recommending targeted campaigns.

[0080] Step 11: Communicate your strategy

[0081] The server notifies the user of the generated marketing and sales strategies via the dashboard, email, and push notifications.

[0082] Step 12: Get feedback

[0083] Users report the results of their campaigns and strategies to the server, specifically by inputting the success rate and KPIs (key performance indicators) of the campaigns.

[0084] Step 13: Retrain the model

[0085] The server uses feedback data obtained from users to retrain the generative AI model to improve its accuracy, and adds new data to continuously improve the model's performance.

[0086] The above are the specific processing steps for carrying out the invention.

[0087] Example 1

[0088] 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."

[0089] Traditional methods for formulating marketing and sales strategies make it difficult to grasp real-time market trends, making it difficult to quickly derive effective strategies. Furthermore, there is a lack of a process for improving the system based on feedback, making it difficult to improve accuracy. Furthermore, existing systems require a great deal of time and effort to preprocess and analyze massive amounts of data, making efficient operation difficult.

[0090] 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.

[0091] In this invention, the server includes means for selecting a data source from a user and accessing an external API or an internal database, means for preprocessing the collected data and filtering out unnecessary information, means for inputting the preprocessed data into a generative AI model and analyzing brand awareness, reputation, and target audience attributes, means for displaying the analysis results on a dashboard in real time and providing them to the user, means for predicting future trends and providing the prediction results, means for generating a marketing and sales strategy using the generative AI model and notifying the user, and means for obtaining user feedback and retraining the model. This makes it possible to grasp market trends in real time and formulate highly accurate marketing and sales strategies.

[0092] A "data source" is a reference from which data is collected, such as an external API or an internal database.

[0093] "Preprocessing" is the process of filtering unnecessary information from collected data and standardizing the data format.

[0094] A "generative AI model" is a model that uses machine learning algorithms to analyze data and generate insights.

[0095] "Brand awareness" is an indicator of how well a particular brand is known among the general public.

[0096] "Reputation" refers to consumer opinions and ratings of a particular brand or product.

[0097] A "target audience" is a group of consumers that is the target of marketing and advertising.

[0098] "Displaying on the dashboard in real time" means that the analysis results are immediately reflected in the user interface so that the user can check them immediately.

[0099] "Predicting future trends" means predicting future market trends and fluctuations in brand awareness based on collected data.

[0100] A "marketing and sales strategy" is a strategic plan for increasing sales of a product or service.

[0101] "User notification" is the process of informing users of generated strategies and important information.

[0102] "Feedback" is any opinion or evaluation provided based on the results of an implemented strategy or campaign.

[0103] "Model retraining" is the re-learning process of using collected feedback data to improve the accuracy of an existing generative AI model.

[0104] This invention is a system that utilizes generative AI to understand brand awareness, reputation, and target audience insights in real time to enhance marketing and sales strategies. This system is composed of multiple servers, terminals, and users.

[0105] System Configuration

[0106] 1. Data Collection Methods

[0107] The server accesses external APIs and internal databases to collect data related to a specific brand. For example, the server uses the Twitter API to collect tweets containing a specific brand name or related hashtags. The server also retrieves past purchase history and customer information from internal databases. This data is obtained from databases such as MongoDB and MySQL.

[0108] 2. Data preprocessing methods

[0109] The server preprocesses the collected data and filters out unnecessary information. Specifically, the server removes spam messages and irrelevant data and standardizes the format of the text data. During this process, the server uses natural language processing techniques to analyze the text data using the Python libraries NLTK and spaCy.

[0110] 3. Data Analysis Methods

[0111] The server inputs the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target audience attributes. At this stage, the server performs sentiment analysis using the Hugging Face Transformer model to classify the data as positive, negative, or neutral. The server also uses a clustering algorithm (e.g., K-means) to generate a target audience profile based on user purchasing behavior and hobbies.

[0112] 4. Real-time feedback methods

[0113] The server displays the analysis results in real time on a dashboard and provides them to the user. For example, the server uses WebSocket technology to continuously update the dashboard's UI components with the latest analysis results. The front-end terminal uses React.js and D3.js to update graphs and charts based on the received data.

[0114] 5. Means of predicting future trends

[0115] The server uses a generative AI model to predict future trends for the brand. Based on time-series data, it uses a Long Short-Term Memory (LSTM) network to predict fluctuations in awareness and reputation over the next few months, and displays the results on a dashboard.

[0116] 6. Strategy Generation Method

[0117] The server uses the generative AI model to generate marketing and sales strategies and notify users. Specifically, the server uses the generative AI model to propose strategies such as "recommend offering a promotional code in the next campaign" based on demand forecasts. These strategies are not only displayed on the dashboard, but also notified to users via email and push notifications using AWS SNS (Simple Notification Service).

[0118] 7. Feedback channels

[0119] Users provide feedback to the server on the results of their campaigns and strategies. They input the results data using their devices and send it to the server. The server then retrains the generative AI model based on the collected feedback data, improving the accuracy of the next analysis and prediction. TensorFlow and PyTorch are used for retraining.

[0120] Specific examples

[0121] For example, a beverage manufacturer might use this system when launching a new product. The server collects tweets related to the "new beverage name" from social media and analyzes its popularity in real time. The server also combines this with past purchasing data to identify that "men in their 30s who love sports" are the primary target audience. The generative AI model then predicts that the product's popularity will increase next month. Based on this information, the system suggests a strategy to the user, such as "running a promotion during a sporting event," and provides feedback on the results to be used next time.

[0122] (Example of a prompt)

[0123] Collect data to analyze your brand awareness and reputation.

[0124] Preprocess the collected data to filter out irrelevant data.

[0125] Analyze the preprocessed data and classify it as positive, negative, or neutral.

[0126] Predict your brand's future reputation and visibility.

[0127] Based on the analysis, suggest a strategy for your next campaign.

[0128] The above is a specific embodiment of the present invention. By using this system, companies can quickly and accurately grasp market trends and implement effective marketing and sales strategies.

[0129] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0130] Step 1: Data source selection and data collection

[0131] 1. Input: The user selects a data source containing brand names and related keywords.

[0132] 2. Processing: The server accesses external APIs and internal databases based on the input information received from the user. Specifically, it uses the Twitter API to collect tweets containing the specified brand name or hashtag. It also retrieves past purchase history and customer information from internal databases (e.g., MongoDB or MySQL).

[0133] 3. Output: The collected data is stored in the server and is ready for the next step.

[0134] Step 2: Data Preprocessing

[0135] 1. Input: Data collected in step 1.

[0136] 2. Processing: The server preprocesses the data, removing spam messages and irrelevant data and standardizing the format of the text data. Specifically, it uses the Python libraries NLTK and spaCy to parse the text data and convert the format.

[0137] 3. Output: Preprocessed and clean data is generated and sent to the next step.

[0138] Step 3: Input and analysis into the generative AI model

[0139] 1. Input: The preprocessed data from step 2.

[0140] 2. Processing: The server inputs this data into a generative AI model. It uses Hugging Face's Transformer model to perform sentiment analysis and classify the data as positive, negative, or neutral. It also uses a clustering algorithm (e.g., K-means) to generate a profile of the target audience.

[0141] 3. Output: The analyzed data is generated and results are obtained regarding awareness, reputation, and target audience attributes.

[0142] Step 4: Real-time feedback

[0143] 1. Input: The analysis results generated in step 3.

[0144] 2. Processing: The server reflects the results in real time on the dashboard. Using WebSocket technology, the latest analysis results are updated sequentially on the dashboard's UI components. The terminal visualizes the data using React.js and D3.js.

[0145] 3. Output: The latest data is displayed in the dashboard that users access.

[0146] Step 5: Predict future trends

[0147] 1. Input: Analysis results and time series data generated in Step 3.

[0148] 2. Processing: The server uses a Long Short-Term Memory (LSTM) network to predict future fluctuations in popularity and reputation.

[0149] 3. Output: The prediction results are generated and displayed on a dashboard.

[0150] Step 6: Strategy Generation

[0151] 1. Input: The analytical and prediction results generated in Step 3 and Step 5.

[0152] 2. Processing: The server uses the generative AI model to generate marketing and sales strategies, such as providing promotional codes for the next campaign.

[0153] 3. Output: The generated strategies are displayed on the dashboard and notified to the user via email and push notifications using AWS SNS (Simple Notification Service).

[0154] Step 7: Get feedback and retrain the model

[0155] 1. Input: The user inputs the results of the campaigns and strategies they have executed using a terminal and sends them to the server.

[0156] 2. Processing: The server uses the collected feedback data to retrain the generative AI model using TensorFlow or PyTorch.

[0157] 3. Output: A retrained AI model is generated, improving the accuracy of the next analysis or prediction.

[0158] (Application example 1)

[0159] 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."

[0160] It has been difficult to accurately understand the effectiveness of traditional advertising campaigns in real time and adjust strategies immediately. It has also been difficult to generate detailed profiles of target users based on their purchasing behavior and hobbies. Furthermore, there has been a lack of efficient means to process large amounts of data obtained from social media and provide sentiment analysis results in real time.

[0161] 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.

[0162] In this invention, the server includes means for selecting information sources from users and accessing external APIs and internal databases, means for preprocessing the collected data and filtering unnecessary information, means for inputting the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target user attributes, means for displaying sentiment analysis results collected from social media on a dashboard in real time and providing them to users, means for predicting future trends and providing the prediction results, means for generating marketing and sales strategies based on the analysis results and prediction results and notifying users, and means for obtaining user feedback and retraining the model. This makes it possible to accurately grasp the effectiveness of advertising campaigns in real time and immediately adjust strategies.

[0163] "User" means any person or entity that uses this system.

[0164] A "source" is a service or system that provides data, such as an external API or an internal database.

[0165] A "generative AI model" is an artificial intelligence model that is trained to take data as input and analyze brand awareness, reputation, and target user attributes.

[0166] "Social media" refers to online platforms such as Twitter where users post and share information.

[0167] "Sentiment analysis" is the process of classifying the sentiment of text data into positive, negative, or neutral.

[0168] A "dashboard" is an interface for displaying analysis results and prediction results in real time.

[0169] A "marketing strategy" is a plan or method aimed at promoting a product or service.

[0170] A "sales strategy" refers to the plans and methods of sales activities aimed at increasing sales.

[0171] "Feedback" refers to information received from users regarding the results of implemented campaigns and strategies.

[0172] "Retraining" is the process of retraining an artificial intelligence model based on feedback data.

[0173] "Future trends" refers to future trends predicted based on time series data and generative AI models.

[0174] This invention is a system that utilizes generative AI models to understand brand awareness, reputation, and target user insights in real time to enhance marketing and sales strategies. The system consists of a server and an application that can be accessed from a smartphone or other device. A specific example is shown below.

[0175] Functional Configuration

[0176] 1. Data Collection Methods

[0177] The server selects the information source specified by the user and accesses external APIs (e.g., Twitter API) or internal databases. The server collects data containing the specified brand name and related hashtags. This data includes social media posts, the user's past purchase history, customer information, etc.

[0178] 2. Data preprocessing methods

[0179] The server preprocesses the collected data and filters out unnecessary information, using natural language processing techniques to remove spam messages and irrelevant data and standardize the format of the text data.

[0180] 3. Data Analysis Methods

[0181] The preprocessed data is input into a generative AI model, and the server analyzes brand awareness, reputation, and target user attributes. A machine learning model is used to perform sentiment analysis and classify the data as positive, negative, or neutral. It also generates target user profiles based on user purchasing behavior and hobbies.

[0182] 4. Real-time feedback methods

[0183] The server displays the analysis results in real time on a dashboard and provides them to users. WebSocket technology is used to continuously update the dashboard UI components with the latest analysis results, allowing users to grasp market trends in real time.

[0184] 5. Means of predicting future trends

[0185] The server uses generative AI models to predict future trends, using time-series data to forecast fluctuations in popularity and reputation over the next few months, and displays the results on a dashboard.

[0186] 6. Strategy Generation Method

[0187] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user of them, specifically proposing the content and timing of promotions to be carried out in the next campaign.

[0188] 7. Feedback channels

[0189] Users provide feedback on the results of their campaigns and strategies to the server, which then uses this feedback to retrain the AI ​​model and improve the system's accuracy, enabling it to provide more effective strategies for the next campaign.

[0190] Examples of specific examples and prompts

[0191] For example, a beverage manufacturer might use this system when running an advertising campaign for a new product. The server collects tweets related to the "new beverage name" from social media and analyzes their popularity in real time. The server also combines this with past purchasing data to identify that "men in their 30s who love sports" are the primary target users. Based on this information, the generative AI model predicts that "product popularity will increase next month" and suggests a strategy to users: "run a promotion during a sporting event." When the results of the campaign are fed back, the AI ​​model is retrained using that data.

[0192] Example prompt sentence:

[0193] """

[0194] Your goal is to write a program that tracks social media sentiment related to a particular brand and analyzes that sentiment. You will use the Twitter API for data collection and natural language processing techniques for data preprocessing and sentiment analysis. You will set up a web server using Flask to display the results in real time.

[0195] """

[0196] Using this system, companies can quickly and accurately grasp market trends and implement effective marketing and sales strategies.

[0197] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0198] Step 1:

[0199] The server selects information sources specified by the user and collects data. Specifically, when the user specifies a specific brand name or related hashtag, the server collects social media posts via the Twitter API. It also accesses an internal database to obtain related purchase history and customer information. The input is the specified brand name and hashtag, and the output is the collected social media posts and customer data.

[0200] Step 2:

[0201] The server preprocesses the collected data and filters out unnecessary information. Specifically, it uses natural language processing techniques to remove spam messages and irrelevant data and standardize the format of the text data. This process involves, for example, using automated scripts to remove URLs and mentions and remove stop words. The input is the collected social media posts and customer data, and the output is the preprocessed data.

[0202] Step 3:

[0203] The preprocessed data is input into a generative AI model to analyze brand awareness, reputation, and target user attributes. The server uses a machine learning model to perform sentiment analysis on the text data and classify the data as positive, negative, or neutral. It also generates a target user profile based on the user's purchasing behavior and hobbies. The input is the preprocessed data, and the output is the sentiment analysis results and the target user profile.

[0204] Step 4:

[0205] The server displays the analysis results in real time on a dashboard and provides them to users. WebSocket technology is used to continuously update the latest analysis results to the dashboard's UI components, allowing users to grasp market trends in real time. The input is the sentiment analysis results and target user profiles, and the output is real-time analysis information displayed on the dashboard.

[0206] Step 5:

[0207] The server uses a generative AI model to predict future trends. Based on time-series data, it predicts fluctuations in popularity and reputation over the next few months and displays the results on a dashboard. Specifically, it uses a predictive algorithm to calculate future data points and visualizes the results. The input is time-series data, and the output is a forecast of future trends.

[0208] Step 6:

[0209] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user. Specifically, it proposes the content and timing of promotions to be carried out in the next campaign. The server sends the generated strategy information to the user via email or push notification. The input is the analysis and prediction results, and the output is a proposed marketing and sales strategy.

[0210] Step 7:

[0211] Users provide feedback to the server on the results of their implemented campaigns and strategies. The server uses this feedback to retrain the AI ​​model and improve the accuracy of the system. For example, a user inputs the effectiveness of a campaign, and the server collects that data and uses it for the next model training. The input is the user's feedback data, and the output is a retrained AI model.

[0212] 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.

[0213] This invention is a system that utilizes generative AI to grasp brand awareness, reputation, and target audience insights in real time, and combines it with an emotion engine that recognizes user emotions to enhance marketing and sales strategies. This system is composed of multiple servers, terminals, and users.

[0214] System Configuration

[0215] 1. Data Collection Methods

[0216] The server accesses external APIs and internal databases to collect data related to a specific brand. For example, the server uses the Twitter API to collect tweets containing the brand name and related hashtags. At the same time, it retrieves past purchase history and customer information from the internal database.

[0217] 2. Data preprocessing methods

[0218] The server preprocesses the collected data and filters out unnecessary information. Specifically, the server removes spam messages and irrelevant data and standardizes the format of the text data. During this process, the server analyzes the text data using natural language processing technology.

[0219] 3. Data Analysis Methods

[0220] The server inputs the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target audience attributes. The server then uses a machine learning model to perform sentiment analysis and classify the data as positive, negative, or neutral. It also generates a target audience profile based on user purchasing behavior and hobbies.

[0221] 4. Real-time feedback methods

[0222] The server displays the analysis results in real time on a dashboard and provides them to the user. For example, the server uses WebSocket technology to continuously update the latest analysis results to the dashboard UI component, allowing users to grasp market trends in real time.

[0223] 5. Means of predicting future trends

[0224] Sarver uses generative AI models to predict future trends for brands. Based on time-series data, it predicts fluctuations in awareness and reputation over the next few months and displays the results on a dashboard. For example, Sarver predicts that brand awareness will increase by 5% next month.

[0225] 6. Strategy Generation Method

[0226] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user. Specifically, the server makes strategy suggestions such as "recommending providing a promotion code in the next campaign." These suggestions are notified to the user not only via the dashboard but also by email and push notifications.

[0227] 7. Feedback channels

[0228] Users provide feedback on the results of their campaigns and strategies to the server. The server uses this feedback to retrain the AI ​​model and improve the accuracy of the system. For example, a user can input the results of a campaign, and the server can collect that data and use it for the next model training.

[0229] 8. Emotional Engine Means

[0230] The server uses an emotion engine to recognize the user's emotions in real time. For example, the server performs emotion analysis based on data acquired from the user's online activities (comments, purchase history, access logs, etc.) to identify the user's current emotional state.

[0231] 9. Personalized Strategy Generation

[0232] The server generates and provides personalized marketing and sales strategies to the user based on the user's emotional state, for example, promoting a specific promotion if the user is in a positive emotional state and providing customer support information if the user is in a negative emotional state.

[0233] Specific examples

[0234] For example, a beverage manufacturer could use this system when launching a new product. The server collects related tweets on the "new beverage name" from social media and analyzes their popularity in real time. The server also combines this with past purchasing data to identify that "sports-loving men in their 30s" are the primary target audience. The generative AI model then predicts that the product's popularity will increase next month. Furthermore, the emotion engine analyzes the emotional state of users from their comments and feedback and provides personalized strategies. For example, if there are a lot of positive reactions, a limited promotional campaign can be promoted, and if there are a lot of negative reactions, immediate customer support can be provided.

[0235] The above is a specific embodiment of the present invention. By using this system, companies can quickly and accurately grasp market trends, implement effective marketing and sales strategies, and even provide personalized responses that take into account the emotional state of users.

[0236] The processing flow will be explained below.

[0237] Step 1: Select a data source

[0238] The server selects a data source to allow it to collect data from an external API (e.g., Twitter API) or an internal database. The server obtains an API key and authentication information to establish access to the external API.

[0239] Step 2: Getting the data

[0240] The server sends a request containing specific keywords or hashtags to retrieve data from selected data sources. Specifically, the server sends an API request to search tweets and posts for the brand name and related keywords.

[0241] Step 3: Preprocessing the data

[0242] The server preprocesses the retrieved data and filters out unnecessary information, specifically filtering out spam messages and normalizing the text (e.g., converting to lowercase and removing extra whitespace). In this step, natural language processing techniques are used to analyze the text data.

[0243] Step 4: Save your data

[0244] The server stores the pre-processed data in an internal database for quick access in subsequent processing steps.

[0245] Step 5: Perform sentiment analysis

[0246] The server uses a generative AI model to perform sentiment analysis on the pre-processed data, for example, taking tweet or review text as input and classifying it as positive, negative, or neutral.

[0247] Step 6: Target Audience Analysis

[0248] The server combines the results of the sentiment analysis with the user's purchasing history to perform a detailed analysis of the target audience's attributes, specifically creating a profile of their age group, gender, hobbies, and preferences.

[0249] Step 7: Provide real-time feedback

[0250] The server displays the analysis results in a dashboard in real time, using WebSocket or API to ensure that data is reflected to users in real time.

[0251] Step 8: Perform future trend predictions

[0252] The server uses a generative AI model to predict future trends for the brand, predicting fluctuations in name recognition and reputation based on time-series data.

[0253] Step 9: Providing prediction results

[0254] The server displays the prediction results on a dashboard and provides them to the user, visualizing them in graphs and charts for easy understanding.

[0255] Step 10: Generate a marketing and sales strategy

[0256] Based on the analysis and prediction results, the server proposes specific marketing and sales strategies to the user, such as providing promotion codes and recommending targeted campaigns.

[0257] Step 11: Communicate your strategy

[0258] The server notifies the user of the generated marketing and sales strategies via the dashboard, email, and push notifications.

[0259] Step 12: Collect user emotion data

[0260] The server collects data on the user's emotional state from their online activities (comments, purchase history, access logs, etc.) and uses an emotion engine to analyze the collected data in real time to identify the user's emotional state.

[0261] Step 13: Individualized strategy generation

[0262] The server generates and provides personalized marketing and sales strategies based on the user's emotional state, for example, promoting new product introductions and promotional campaigns to users in a positive emotional state, and providing customer support information to users in a negative emotional state.

[0263] Step 14: Getting feedback

[0264] Users report the results of their campaigns and strategies to the server, specifically by inputting the success rate and KPIs (key performance indicators) of the campaigns.

[0265] Step 15: Retrain the model

[0266] The server uses feedback data obtained from users to retrain the generative AI model and emotion engine to improve accuracy, and adds new data to continuously improve the model's performance.

[0267] The above are the specific processing steps for carrying out the invention.

[0268] Example 2

[0269] 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."

[0270] Traditional marketing and sales strategies struggle to collect and analyze large amounts of data in real time to generate effective strategies. It's also difficult to provide personalized responses that take into account the user's emotional state. Furthermore, there's a lack of a mechanism for continuously improving models based on feedback, limiting the accuracy of the system.

[0271] 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: means for selecting information sources from users and accessing external APIs and internal storage devices; means for preprocessing the collected information and filtering unnecessary information; means for inputting the preprocessed information into a generative AI model and analyzing brand awareness, reputation, and target audience attributes; means for displaying the analysis results on a display device in real time and providing them to the user; means for predicting future trends and providing the prediction results; means for generating sales promotion and sales tactics based on the analysis and prediction results and notifying the user; means for obtaining user feedback and retraining the model; means for recognizing user emotions in real time using an emotion engine; and means for generating personalized strategies based on the user's emotional state and providing them to the user. This enables rapid and accurate understanding of market trends and individualized responses that take the user's emotional state into consideration. Furthermore, by continuously collecting feedback and improving the model, the accuracy of the system can be improved.

[0272] "User" refers to the entity that uses this system to execute marketing and sales strategies.

[0273] "Source" refers to an external API or internal storage device from which data is obtained.

[0274] "External API" refers to an application program interface for obtaining data from external services provided on the Internet.

[0275] "Internal storage device" refers to a database that stores past purchase history, customer information, etc., managed within a company.

[0276] "Collected Information" refers to data obtained by the Server from sources designated by the User.

[0277] "Preprocessing" refers to the process of filtering unnecessary information from collected information and standardizing the format of text data.

[0278] A "generative AI model" refers to an artificial intelligence model that can analyze data and make predictions in a generative way.

[0279] "Brand awareness" refers to an indicator of how familiar consumers are with a particular brand.

[0280] "Reputation" refers to the general evaluation or opinion of a brand or product.

[0281] A "target audience" refers to a group of customers who may be interested in a particular product or service.

[0282] "Display Device" refers to a device (e.g., dashboard, monitor) for visually displaying collected and analyzed data.

[0283] "Future trends" refers to predicted changes and trends in the market or brand in the future.

[0284] "Sales promotion" refers to the strategies and measures taken to promote a particular product or service in the market.

[0285] "Sales tactics" refer to the specific approaches and methods used to effectively sell a company's products or services.

[0286] "Feedback" refers to information provided to the server regarding the results of campaigns and strategies implemented by the user.

[0287] "Model retraining" refers to the process of improving a generative AI model based on feedback to improve its performance.

[0288] "Emotion engine" refers to a system component for identifying and analyzing user emotions in real time.

[0289] An "individualized strategy" refers to a strategy that provides optimal marketing and sales tactics to specific users based on their emotional state or specific attributes.

[0290] The system uses generative AI to understand brand awareness, reputation, and target audience insights in real time, combined with an emotion engine that recognizes user emotions to enhance marketing and sales strategies. The entire system is comprised of multiple servers, terminals, and users.

[0291] Hardware and Software Configuration

[0292] The server is primarily responsible for data collection, preprocessing, analysis, real-time feedback, prediction, strategy generation, and emotion recognition. The specific hardware used includes a network interface for processing API requests, a database server for storing data, and a high-performance GPU for running the generative AI model. The software includes the Twitter API, an SQL database, natural language processing libraries (e.g., NLTK and Spacy), and machine learning frameworks (e.g., TensorFlow and PyTorch).

[0293] The terminal is a computer or smartphone that users use to access the system and check analysis results and strategic proposals. The terminal connects to the server via a web browser or dedicated application.

[0294] Examples of data collection

[0295] The server collects data related to a specific brand by accessing external APIs (such as the Twitter API) or an internal database. For example, the server collects tweets containing "new drink name" and other related hashtags. At the same time, the server retrieves past purchase history and customer information from the internal database using SQL queries.

[0296] Example prompt sentence:

[0297] Get the latest tweets related to "New Drink Names".

[0298] Specific examples of data preprocessing

[0299] The server preprocesses the collected data, filtering out unnecessary information, removing spam messages and irrelevant data, and standardizing the format of text data. Specifically, it uses natural language processing technologies (e.g., NLTK and Spacy) to clean up and standardize the text format of tweets.

[0300] Example prompt sentence:

[0301] Please remove spam messages from your data and standardize the format.

[0302] Specific examples of data analysis

[0303] The server inputs the preprocessed data into a generative AI model (e.g., BERT or GPT-3) to analyze brand awareness, reputation, and target audience attributes. It also uses a machine learning model to perform sentiment analysis and classify the data as positive, negative, or neutral. It also generates a target audience profile based on user purchasing behavior and hobbies.

[0304] Example prompt sentence:

[0305] Analyze brand sentiment and categorize sentiment based on pre-processed data.

[0306] Examples of real-time feedback

[0307] The server displays the analysis results in real time on a dashboard and provides them to users. This uses WebSocket technology to continuously update the latest analysis results on the dashboard UI component, allowing users to grasp market trends in real time.

[0308] Example prompt sentence:

[0309] View the latest analytics on your dashboard.

[0310] Specific examples of future trend predictions

[0311] The server uses generative AI models to predict future trends for the brand. For example, the server creates a predictive model based on time-series data to forecast fluctuations in brand awareness and reputation over the next few months. The results are displayed on a dashboard.

[0312] Example prompt sentence:

[0313] Predict your brand awareness for the next three months.

[0314] Example of strategy generation

[0315] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user. Specifically, it makes strategic suggestions such as "We recommend offering a promotion code in the next campaign." These suggestions are not only displayed on the dashboard, but also sent to the user via email and push notifications.

[0316] Example prompt sentence:

[0317] Generate recommended strategies for your next campaign.

[0318] Specific examples of feedback

[0319] Users provide feedback on the results of their campaigns and strategies to the server. The server receives this feedback and retrains the AI ​​model to improve the accuracy of the system. For example, users input campaign results through a dashboard or dedicated app, and the server collects the data and uses it for the next model training.

[0320] Example prompt sentence:

[0321] Save the campaign results as feedback.

[0322] Examples of emotion engines

[0323] The server uses an emotion engine to recognize users' emotions in real time. It performs emotion analysis based on the user's online activities (comments, purchase history, access logs, etc.) to identify their current emotional state. For example, it calculates an emotion score based on the user's comments and purchase history and classifies them as positive, negative, or neutral.

[0324] Example prompt sentence:

[0325] Perform sentiment analysis on user comments.

[0326] Examples of personalized strategies

[0327] The server generates and provides personalized marketing and sales strategies based on the user's emotional state, promoting specific promotions in the case of a positive emotional state and providing customer support information in the case of a negative emotional state.

[0328] Example prompt sentence:

[0329] Generate strategies based on the user's emotional state.

[0330] Using this system, companies can quickly and accurately grasp market trends, implement effective marketing and sales strategies, and even provide personalized responses that take into account the user's emotional state.

[0331] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0332] Step 1: Data collection

[0333] The server collects data from information sources. Specifically, the server accesses an external API (e.g., Twitter API) and retrieves tweets containing specified keywords (e.g., "new beverage name"). The server also retrieves past purchase history and customer information from its internal storage device using SQL queries. The input here is the information source and keywords and conditions to be collected, and the output is the collected raw data.

[0334] Specific behavior:

[0335] The server periodically issues requests to the Twitter API and parses the API responses to retrieve data.

[0336] The server extracts the required information from the internal storage using SQL queries.

[0337] Step 2: Data Preprocessing

[0338] The server preprocesses the collected data and filters out unnecessary information. Specifically, the server uses natural language processing techniques (e.g., NLTK or Spacy) to clean up the tweet text data and standardize the format. The input here is the collected raw data, and the output is the preprocessed clean data.

[0339] Specific behavior:

[0340] The server cleans the text data, removing spam messages and irrelevant data.

[0341] The server uses regular expressions to remove certain patterns and converts the text data into a unified format.

[0342] Step 3: Data analysis

[0343] The server then inputs the preprocessed data into a generative AI model (e.g., BERT or GPT-3) to analyze brand awareness, reputation, and target audience attributes. The input is the preprocessed clean data, and the output is the analysis results. The server also uses a machine learning model to perform sentiment analysis and classify the data as positive, negative, or neutral.

[0344] Specific behavior:

[0345] The server inputs the preprocessed text into a generative AI model to obtain emotion scores and attribute information.

[0346] The server uses a sentiment analysis model to classify tweets and assign them to positive, negative, or neutral sentiment categories.

[0347] Step 4: Real-time feedback

[0348] The server displays the analysis results in real time on a display device and provides them to the user. The input here is the results of the data analysis, and the output is dashboard data that is updated in real time.

[0349] Specific behavior:

[0350] The server pushes data to the dashboard over a WebSocket connection.

[0351] The dashboard UI components receive the latest data and present it visually to the user.

[0352] Step 5: Predict future trends

[0353] The server uses a generative AI model to predict future trends for the brand, where the input is past data and current analysis results, and the output is predicted future trend data.

[0354] Specific behavior:

[0355] The server predicts the data using a time series prediction model (e.g., ARIMA or LSTM).

[0356] The server sends the prediction results to a dashboard for display to the user.

[0357] Step 6: Strategy Generation

[0358] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user, where the inputs are the analysis and prediction results and the output is the generated strategy proposals.

[0359] Specific behavior:

[0360] The server generates a strategy template based on the analysis and prediction results.

[0361] The server sends the generated strategy to the user via email or push notification.

[0362] Step 7: Gather feedback

[0363] The results of the campaigns and strategies executed by the user are provided as feedback to the server. The input here is the results of the campaigns and strategies, and the output is the collected feedback data. The server uses this feedback to retrain the AI ​​model.

[0364] Specific behavior:

[0365] Users enter campaign results through a dashboard or dedicated app.

[0366] The server collects the feedback data and uses it for the next model training.

[0367] Step 8: Emotion Recognition

[0368] The server uses an emotion engine to recognize the user's emotions in real time, where the input is the user's online activity data and the output is the recognized emotional state.

[0369] Specific behavior:

[0370] The server analyzes the user's comments and purchase history and calculates an emotion score.

[0371] The server appropriately tags and stores the emotional state.

[0372] Step 9: Personalized strategy generation

[0373] The server generates personalized marketing and sales strategies based on the user's emotional state and provides them to the user, where the input is the user's emotional state and the output is personalized strategy proposals.

[0374] Specific behavior:

[0375] The server classifies the user's emotional state and selects an appropriate strategy template.

[0376] The server generates a personalized strategy and notifies the user via email or push notification.

[0377] (Application example 2)

[0378] 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."

[0379] Product quality control is a critical issue in modern factory operations. Mass-produced products, in particular, face a high risk of quality variation, making quality control time-consuming and costly. Traditional quality control systems struggle to identify problems in real time and propose rapid solutions. Systems that can predict future quality trends and propose appropriate improvement measures are also lacking. This creates a demand for advanced systems that can maintain high levels of product quality while improving factory efficiency.

[0380] 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.

[0381] In this invention, the server includes: means for collecting product quality data from sensors and camera devices installed in the factory; means for preprocessing the collected product quality data and filtering incomplete or abnormal data; means for inputting the preprocessed quality data into a generative AI model to monitor, evaluate, and identify problems in real time; means for displaying analysis results in real time on a dashboard for factory workers and immediately notifying them if a problem occurs; means for predicting future quality trends based on past quality data and providing the prediction results to workers; and means for proposing optimal product inspection methods and manufacturing processes based on the analysis and prediction results. This enables highly accurate real-time monitoring and rapid response, not only enabling efficient product quality control but also enabling proactive improvement measures to be implemented based on future trend predictions.

[0382] A "data source" is a source of information from which information is obtained, such as an external API or an internal database.

[0383] "External API" refers to an interface to external services or data provided over the Internet.

[0384] An "internal database" is a system that stores data managed within an organization.

[0385] "Preprocessing" refers to the process of converting raw data into a form that is easier to analyze, such as removing noise and standardizing the format.

[0386] A "generative AI model" is a model that uses generative adversarial networks (GANs) or other AI algorithms to analyze data and generate new information.

[0387] A "brand" is a collective term for the elements that create a particular image and reputation for a company or product.

[0388] "Awareness" is a measure of how many people know about a brand or product.

[0389] "Reputation" refers to the evaluation and feelings that consumers and the market have about a brand or product.

[0390] "Target audience" refers to a specific group of consumers that you are targeting with marketing and advertising.

[0391] "Attributes" refer to characteristics of the target audience, such as gender, age, and hobbies.

[0392] "Real-time" refers to the state in which data is acquired and processed almost immediately.

[0393] A "dashboard" is an interface that visually displays data analysis results and important information.

[0394] "Future trends" refers to future tendencies and developments predicted by analyzing past and present data.

[0395] "Feedback" refers to information used to adjust future actions based on the results and reactions of implemented measures and strategies.

[0396] "Factory" refers to a place where products are produced or processed.

[0397] A "sensor" is a device that senses physical conditions or data and converts them into electrical signals.

[0398] A "camera device" is a device that captures video and converts it into digital data.

[0399] "Quality Data" refers to data relating to the characteristics and performance of a product.

[0400] "Incomplete data" refers to data that is missing or missing.

[0401] "Abnormal data" refers to data that exhibits unusual values ​​or patterns that fall outside of normal ranges.

[0402] "Monitoring" means constantly checking a specific object or condition.

[0403] "Evaluation" means judging the quality or performance of something.

[0404] "Notification" refers to the immediate sharing of specific information or results.

[0405] "Quality Trends" refers to trends or fluctuations in product quality.

[0406] "Improvement measures" refer to methods and means for solving current problems and improving the situation.

[0407] "Product inspection methods" refer to specific methods and techniques for checking product quality.

[0408] "Manufacturing process" refers to the series of steps taken to create a product from raw materials.

[0409] This invention provides a system for streamlining factory product quality control and continuously providing high-quality products. This system utilizes various sensors and camera devices installed in the factory to collect and analyze product quality data in real time. It also proposes quick countermeasures based on the analysis results and predicts future quality trends, thereby improving product quality.

[0410] System Configuration

[0411] The server operates in the following specific steps:

[0412] Data collection methods

[0413] The server collects product quality data using sensors and camera devices installed in the factory, including various quality indicators such as product dimensions, weight, and appearance. Specifically, it uses IoT devices and communication modules to acquire data from the various sensors and cameras via a network.

[0414] Data preprocessing measures

[0415] The server preprocesses the collected data, filtering out incomplete or abnormal data, and utilizes open-source natural language processing libraries (e.g., NLTK) and image processing libraries (e.g., OpenCV) to remove noise from the data and convert it into a unified format.

[0416] Data Analysis Methods

[0417] The preprocessed data is input into a generative AI model. The server uses the generative AI model to monitor product quality in real time and identify problems. A sentiment engine (e.g., Sentiment Analyzer) is also used for analysis, analyzing quality data and feedback logs from workers to identify the cause of quality issues.

[0418] Real-time feedback tools

[0419] The server displays the analysis results in real time on a dashboard for factory workers and immediately notifies them if any problems occur. WebSocket technology is used to continuously update the dashboard's UI components with the latest information, allowing workers to immediately detect problems and take action.

[0420] A means of predicting future trends

[0421] The server predicts future quality trends based on past quality data. To do this, it uses a time series analysis algorithm to predict future fluctuations in product quality. The analysis results are displayed on a dashboard and provided to workers.

[0422] Strategy Generation Method

[0423] Based on the analysis and prediction results, the server proposes optimal product inspection methods and manufacturing processes, allowing factory workers to quickly implement appropriate improvement measures.

[0424] Specific examples

[0425] For example, this system is used in a factory to control the quality of steel products. The server collects data on product dimensions, weight, and surface condition from sensors and camera devices, and evaluates the quality in real time. The preprocessed data is input into a generative AI model to detect quality anomalies. The detection results are displayed on a dashboard and workers are immediately notified. Furthermore, future quality trends are predicted based on past data, and if a prediction is made that the quality of steel products will improve with a certain probability next month, specific improvement measures are suggested to workers.

[0426] Prompt Sentence Examples

[0427] Input product quality data and previous day's manufacturing process data into a generative AI model and suggest your next quality control strategy. For example, are there any trends like the following?

[0428] By using the system of the present invention, not only can factory quality control be carried out efficiently, but proactive improvement measures can also be implemented based on predictions of future trends.

[0429] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0430] Step 1: Data collection

[0431] The server collects product quality data from sensors and camera devices installed in the factory. The sensors acquire physical data such as product dimensions, weight, and temperature, while the camera devices acquire images of the product's appearance. The input is sensor and camera data, and the output is the collected results of this data. Specifically, the server accesses the API of each device, requests data, and acquires it.

[0432] Step 2: Data Preprocessing

[0433] The server preprocesses the collected data and filters out incomplete or abnormal data. Specifically, it performs missing value imputation, outlier detection and removal, and data format standardization. The input is the data collected in step 1, and the output is quality data that has been filtered and cleaned. Specifically, it performs these processes using a natural language processing library (NLTK) and an image processing library (OpenCV).

[0434] Step 3: Data analysis

[0435] The server inputs the preprocessed data into the generative AI model to monitor, evaluate, and identify problems in real time with product quality. The input is preprocessed quality data, and the output is the quality evaluation results. Specifically, the server inputs data into the generative AI model and uses a sentiment engine (Sentiment Analyzer) to evaluate the quality.

[0436] Step 4: Real-time feedback

[0437] The server displays the analysis results in real time on a dashboard for factory workers and immediately notifies them if any problems occur. The dashboard is updated using WebSocket technology. The input is the quality evaluation results, and the output is real-time notifications to workers and the dashboard display. Specifically, the UI components are updated via the WebSocket connection.

[0438] Step 5: Predict future trends

[0439] The server predicts future quality trends based on past quality data. It uses a time series analysis algorithm to predict future quality fluctuations. The input is past and current quality data, and the output is the future quality trend prediction result. Specifically, it performs predictive analysis using a time series analysis algorithm.

[0440] Step 6: Strategy Generation

[0441] The server proposes optimal product inspection methods and manufacturing processes based on the analysis and prediction results. The inputs are prediction results and quality assessment results, and the output is improvement proposals. Specifically, the server notifies the dashboard and workers of the optimal strategy derived by the generative AI model.

[0442] Step 7: Get feedback and retrain

[0443] The server receives feedback on the implemented improvements and retrains the generative AI model. The input is feedback data from the workers, and the output is the retrained generative AI model. Specifically, the server uses the received feedback data to update the AI ​​model's parameters and improve its accuracy.

[0444] 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.

[0445] 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.

[0446] 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.

[0447] [Second embodiment]

[0448] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0449] 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.

[0450] 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).

[0451] 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.

[0452] 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.

[0453] 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).

[0454] 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.

[0455] 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.

[0456] 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.

[0457] 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.

[0458] 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.

[0459] 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."

[0460] This invention is a system that utilizes generative AI to understand brand awareness, reputation, and target audience insights in real time to enhance marketing and sales strategies. This system is composed of multiple servers, terminals, and users.

[0461] System Configuration

[0462] 1. Data Collection Methods

[0463] The server accesses external APIs and internal databases to collect data related to a specific brand. For example, the server uses the Twitter API to collect tweets containing the brand name and related hashtags. At the same time, it retrieves past purchase history and customer information from the internal database.

[0464] 2. Data preprocessing methods

[0465] The server preprocesses the collected data and filters out unnecessary information. Specifically, the server removes spam messages and irrelevant data and standardizes the format of the text data. During this process, the server analyzes the text data using natural language processing technology.

[0466] 3. Data Analysis Methods

[0467] The server inputs the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target audience attributes. At this stage, the server uses a machine learning model to perform sentiment analysis and classify the data as positive, negative, or neutral. It also generates a target audience profile based on user purchasing behavior and hobbies.

[0468] 4. Real-time feedback methods

[0469] The server displays the analysis results in real time on a dashboard and provides them to the user. For example, the server uses WebSocket technology to continuously update the latest analysis results to the dashboard UI component, allowing users to grasp market trends in real time.

[0470] 5. Means of predicting future trends

[0471] Sarver uses generative AI models to predict future trends for brands. Based on time-series data, it predicts fluctuations in awareness and reputation over the next few months and displays the results on a dashboard. For example, Sarver predicts that brand awareness will increase by 5% next month.

[0472] 6. Strategy Generation Method

[0473] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user. Specifically, the server makes strategy suggestions such as "recommending providing a promotion code in the next campaign." These suggestions are notified to the user not only via the dashboard but also by email and push notifications.

[0474] 7. Feedback channels

[0475] Users provide feedback on the results of their campaigns and strategies to the server. The server uses this feedback to retrain the AI ​​model and improve the accuracy of the system. For example, a user can input the results of a campaign, and the server can collect that data and use it for the next model training.

[0476] Specific examples

[0477] For example, a beverage manufacturer might use this system when launching a new product. The server collects tweets related to the "new beverage name" from social media and analyzes its popularity in real time. The server also combines this with past purchasing data to identify that "men in their 30s who love sports" are the primary target audience. The generative AI model then predicts that the product's popularity will increase next month. Based on this information, the system suggests a strategy to the user, such as "running a promotion during a sporting event," and provides feedback on the results to be used next time.

[0478] The above is a specific embodiment of the present invention. By using this system, companies can quickly and accurately grasp market trends and implement effective marketing and sales strategies.

[0479] The processing flow will be explained below.

[0480] Step 1: Select a data source

[0481] A data source is selected so that the server can collect data from an external API (e.g., Twitter API) or an internal database. The server obtains an API key and authentication information to establish access to the external API.

[0482] Step 2: Getting the data

[0483] The server sends a request containing specific keywords or hashtags to retrieve data from selected data sources. Specifically, the server sends an API request to search tweets and posts for the brand name and related keywords.

[0484] Step 3: Preprocessing the data

[0485] The server preprocesses the retrieved data and filters out unnecessary information, specifically filtering out spam messages and normalizing the text (e.g., converting to lowercase and removing extra whitespace). In this step, natural language processing techniques are used to analyze the text data.

[0486] Step 4: Save your data

[0487] The server stores the pre-processed data in an internal database for quick access in subsequent processing steps.

[0488] Step 5: Perform sentiment analysis

[0489] The server uses a generative AI model to perform sentiment analysis on the pre-processed data, for example, taking tweet or review text as input and classifying it as positive, negative, or neutral.

[0490] Step 6: Target Audience Analysis

[0491] The server combines the results of the sentiment analysis with the user's purchasing history to perform a detailed analysis of the target audience's attributes, specifically creating a profile of their age group, gender, hobbies, and preferences.

[0492] Step 7: Provide real-time feedback

[0493] The server displays the analysis results in a dashboard in real time, using WebSocket or API to ensure that data is reflected to users in real time.

[0494] Step 8: Perform future trend predictions

[0495] The server uses a generative AI model to predict future trends for the brand, predicting fluctuations in name recognition and reputation based on time-series data.

[0496] Step 9: Providing prediction results

[0497] The server displays the prediction results on a dashboard and provides them to the user, visualizing them in graphs and charts for easy understanding.

[0498] Step 10: Generate a marketing and sales strategy

[0499] Based on the analysis and prediction results, the server proposes specific marketing and sales strategies to the user, such as providing promotion codes and recommending targeted campaigns.

[0500] Step 11: Communicate your strategy

[0501] The server notifies the user of the generated marketing and sales strategies via the dashboard, email, and push notifications.

[0502] Step 12: Get feedback

[0503] Users report the results of their campaigns and strategies to the server, specifically by inputting the success rate and KPIs (key performance indicators) of the campaigns.

[0504] Step 13: Retrain the model

[0505] The server uses feedback data obtained from users to retrain the generative AI model to improve its accuracy, and adds new data to continuously improve the model's performance.

[0506] The above are the specific processing steps for carrying out the invention.

[0507] Example 1

[0508] 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."

[0509] Traditional methods for formulating marketing and sales strategies make it difficult to grasp real-time market trends, making it difficult to quickly derive effective strategies. Furthermore, there is a lack of a process for improving the system based on feedback, making it difficult to improve accuracy. Furthermore, existing systems require a great deal of time and effort to preprocess and analyze massive amounts of data, making efficient operation difficult.

[0510] 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.

[0511] In this invention, the server includes means for selecting a data source from a user and accessing an external API or an internal database, means for preprocessing the collected data and filtering out unnecessary information, means for inputting the preprocessed data into a generative AI model and analyzing brand awareness, reputation, and target audience attributes, means for displaying the analysis results on a dashboard in real time and providing them to the user, means for predicting future trends and providing the prediction results, means for generating a marketing and sales strategy using the generative AI model and notifying the user, and means for obtaining user feedback and retraining the model. This makes it possible to grasp market trends in real time and formulate highly accurate marketing and sales strategies.

[0512] A "data source" is a reference from which data is collected, such as an external API or an internal database.

[0513] "Preprocessing" is the process of filtering unnecessary information from collected data and standardizing the data format.

[0514] A "generative AI model" is a model that uses machine learning algorithms to analyze data and generate insights.

[0515] "Brand awareness" is an indicator of how well a particular brand is known among the general public.

[0516] "Reputation" refers to consumer opinions and ratings of a particular brand or product.

[0517] A "target audience" is a group of consumers that is the target of marketing and advertising.

[0518] "Displaying on the dashboard in real time" means that the analysis results are immediately reflected in the user interface so that the user can check them immediately.

[0519] "Predicting future trends" means predicting future market trends and fluctuations in brand awareness based on collected data.

[0520] A "marketing and sales strategy" is a strategic plan for increasing sales of a product or service.

[0521] "User notification" is the process of informing users of generated strategies and important information.

[0522] "Feedback" is any opinion or evaluation provided based on the results of an implemented strategy or campaign.

[0523] "Model retraining" is the re-learning process of using collected feedback data to improve the accuracy of an existing generative AI model.

[0524] This invention is a system that utilizes generative AI to understand brand awareness, reputation, and target audience insights in real time to enhance marketing and sales strategies. This system is composed of multiple servers, terminals, and users.

[0525] System Configuration

[0526] 1. Data Collection Methods

[0527] The server accesses external APIs and internal databases to collect data related to a specific brand. For example, the server uses the Twitter API to collect tweets containing a specific brand name or related hashtags. The server also retrieves past purchase history and customer information from internal databases. This data is obtained from databases such as MongoDB and MySQL.

[0528] 2. Data preprocessing methods

[0529] The server preprocesses the collected data and filters out unnecessary information. Specifically, the server removes spam messages and irrelevant data and standardizes the format of the text data. During this process, the server uses natural language processing techniques to analyze the text data using the Python libraries NLTK and spaCy.

[0530] 3. Data Analysis Methods

[0531] The server inputs the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target audience attributes. At this stage, the server performs sentiment analysis using the Hugging Face Transformer model to classify the data as positive, negative, or neutral. The server also uses a clustering algorithm (e.g., K-means) to generate a target audience profile based on user purchasing behavior and hobbies.

[0532] 4. Real-time feedback methods

[0533] The server displays the analysis results in real time on a dashboard and provides them to the user. For example, the server uses WebSocket technology to continuously update the dashboard's UI components with the latest analysis results. The front-end terminal uses React.js and D3.js to update graphs and charts based on the received data.

[0534] 5. Means of predicting future trends

[0535] The server uses a generative AI model to predict future trends for the brand. Based on time-series data, it uses a Long Short-Term Memory (LSTM) network to predict fluctuations in awareness and reputation over the next few months, and displays the results on a dashboard.

[0536] 6. Strategy Generation Method

[0537] The server uses the generative AI model to generate marketing and sales strategies and notify users. Specifically, the server uses the generative AI model to propose strategies such as "recommend offering a promotional code in the next campaign" based on demand forecasts. These strategies are not only displayed on the dashboard, but also notified to users via email and push notifications using AWS SNS (Simple Notification Service).

[0538] 7. Feedback channels

[0539] Users provide feedback to the server on the results of their campaigns and strategies. They input the results data using their devices and send it to the server. The server then retrains the generative AI model based on the collected feedback data, improving the accuracy of the next analysis and prediction. TensorFlow and PyTorch are used for retraining.

[0540] Specific examples

[0541] For example, a beverage manufacturer might use this system when launching a new product. The server collects tweets related to the "new beverage name" from social media and analyzes its popularity in real time. The server also combines this with past purchasing data to identify that "men in their 30s who love sports" are the primary target audience. The generative AI model then predicts that the product's popularity will increase next month. Based on this information, the system suggests a strategy to the user, such as "running a promotion during a sporting event," and provides feedback on the results to be used next time.

[0542] (Example of a prompt)

[0543] Collect data to analyze your brand awareness and reputation.

[0544] Preprocess the collected data to filter out irrelevant data.

[0545] Analyze the preprocessed data and classify it as positive, negative, or neutral.

[0546] Predict your brand's future reputation and visibility.

[0547] Based on the analysis, suggest a strategy for your next campaign.

[0548] The above is a specific embodiment of the present invention. By using this system, companies can quickly and accurately grasp market trends and implement effective marketing and sales strategies.

[0549] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0550] Step 1: Data source selection and data collection

[0551] 1. Input: The user selects a data source containing brand names and related keywords.

[0552] 2. Processing: The server accesses external APIs and internal databases based on the input information received from the user. Specifically, it uses the Twitter API to collect tweets containing the specified brand name or hashtag. It also retrieves past purchase history and customer information from internal databases (e.g., MongoDB or MySQL).

[0553] 3. Output: The collected data is stored in the server and is ready for the next step.

[0554] Step 2: Data Preprocessing

[0555] 1. Input: Data collected in step 1.

[0556] 2. Processing: The server preprocesses the data, removing spam messages and irrelevant data and standardizing the format of the text data. Specifically, it uses the Python libraries NLTK and spaCy to parse the text data and convert the format.

[0557] 3. Output: Preprocessed and clean data is generated and sent to the next step.

[0558] Step 3: Input and analysis into the generative AI model

[0559] 1. Input: The preprocessed data from step 2.

[0560] 2. Processing: The server inputs this data into a generative AI model. It uses Hugging Face's Transformer model to perform sentiment analysis and classify the data as positive, negative, or neutral. It also uses a clustering algorithm (e.g., K-means) to generate a profile of the target audience.

[0561] 3. Output: The analyzed data is generated and results are obtained regarding awareness, reputation, and target audience attributes.

[0562] Step 4: Real-time feedback

[0563] 1. Input: The analysis results generated in step 3.

[0564] 2. Processing: The server reflects the results in real time on the dashboard. Using WebSocket technology, the latest analysis results are updated sequentially on the dashboard's UI components. The terminal visualizes the data using React.js and D3.js.

[0565] 3. Output: The latest data is displayed in the dashboard that users access.

[0566] Step 5: Predict future trends

[0567] 1. Input: Analysis results and time series data generated in Step 3.

[0568] 2. Processing: The server uses a Long Short-Term Memory (LSTM) network to predict future fluctuations in popularity and reputation.

[0569] 3. Output: The prediction results are generated and displayed on a dashboard.

[0570] Step 6: Strategy Generation

[0571] 1. Input: The analytical and prediction results generated in Step 3 and Step 5.

[0572] 2. Processing: The server uses the generative AI model to generate marketing and sales strategies, such as providing promotional codes for the next campaign.

[0573] 3. Output: The generated strategies are displayed on the dashboard and notified to the user via email and push notifications using AWS SNS (Simple Notification Service).

[0574] Step 7: Get feedback and retrain the model

[0575] 1. Input: The user inputs the results of the campaigns and strategies they have executed using a terminal and sends them to the server.

[0576] 2. Processing: The server uses the collected feedback data to retrain the generative AI model using TensorFlow or PyTorch.

[0577] 3. Output: A retrained AI model is generated, improving the accuracy of the next analysis or prediction.

[0578] (Application example 1)

[0579] 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."

[0580] It has been difficult to accurately understand the effectiveness of traditional advertising campaigns in real time and adjust strategies immediately. It has also been difficult to generate detailed profiles of target users based on their purchasing behavior and hobbies. Furthermore, there has been a lack of efficient means to process large amounts of data obtained from social media and provide sentiment analysis results in real time.

[0581] 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.

[0582] In this invention, the server includes means for selecting information sources from users and accessing external APIs and internal databases, means for preprocessing the collected data and filtering unnecessary information, means for inputting the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target user attributes, means for displaying sentiment analysis results collected from social media on a dashboard in real time and providing them to users, means for predicting future trends and providing the prediction results, means for generating marketing and sales strategies based on the analysis results and prediction results and notifying users, and means for obtaining user feedback and retraining the model. This makes it possible to accurately grasp the effectiveness of advertising campaigns in real time and immediately adjust strategies.

[0583] "User" means any person or entity that uses this system.

[0584] A "source" is a service or system that provides data, such as an external API or an internal database.

[0585] A "generative AI model" is an artificial intelligence model that is trained to take data as input and analyze brand awareness, reputation, and target user attributes.

[0586] "Social media" refers to online platforms such as Twitter where users post and share information.

[0587] "Sentiment analysis" is the process of classifying the sentiment of text data into positive, negative, or neutral.

[0588] A "dashboard" is an interface for displaying analysis results and prediction results in real time.

[0589] A "marketing strategy" is a plan or method aimed at promoting a product or service.

[0590] A "sales strategy" refers to the plans and methods of sales activities aimed at increasing sales.

[0591] "Feedback" refers to information received from users regarding the results of implemented campaigns and strategies.

[0592] "Retraining" is the process of retraining an artificial intelligence model based on feedback data.

[0593] "Future trends" refers to future trends predicted based on time series data and generative AI models.

[0594] This invention is a system that utilizes generative AI models to understand brand awareness, reputation, and target user insights in real time to enhance marketing and sales strategies. The system consists of a server and an application that can be accessed from a smartphone or other device. A specific example is shown below.

[0595] Functional Configuration

[0596] 1. Data Collection Methods

[0597] The server selects the information source specified by the user and accesses external APIs (e.g., Twitter API) or internal databases. The server collects data containing the specified brand name and related hashtags. This data includes social media posts, the user's past purchase history, customer information, etc.

[0598] 2. Data preprocessing methods

[0599] The server preprocesses the collected data and filters out unnecessary information, using natural language processing techniques to remove spam messages and irrelevant data and standardize the format of the text data.

[0600] 3. Data Analysis Methods

[0601] The preprocessed data is input into a generative AI model, and the server analyzes brand awareness, reputation, and target user attributes. A machine learning model is used to perform sentiment analysis and classify the data as positive, negative, or neutral. It also generates target user profiles based on user purchasing behavior and hobbies.

[0602] 4. Real-time feedback methods

[0603] The server displays the analysis results in real time on a dashboard and provides them to users. WebSocket technology is used to continuously update the dashboard UI components with the latest analysis results, allowing users to grasp market trends in real time.

[0604] 5. Means of predicting future trends

[0605] The server uses generative AI models to predict future trends, using time-series data to forecast fluctuations in popularity and reputation over the next few months, and displays the results on a dashboard.

[0606] 6. Strategy Generation Method

[0607] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user of them, specifically proposing the content and timing of promotions to be carried out in the next campaign.

[0608] 7. Feedback channels

[0609] Users provide feedback on the results of their campaigns and strategies to the server, which then uses this feedback to retrain the AI ​​model and improve the system's accuracy, enabling it to provide more effective strategies for the next campaign.

[0610] Examples of specific examples and prompts

[0611] For example, a beverage manufacturer might use this system when running an advertising campaign for a new product. The server collects tweets related to the "new beverage name" from social media and analyzes their popularity in real time. The server also combines this with past purchasing data to identify that "men in their 30s who love sports" are the primary target users. Based on this information, the generative AI model predicts that "product popularity will increase next month" and suggests a strategy to users: "run a promotion during a sporting event." When the results of the campaign are fed back, the AI ​​model is retrained using that data.

[0612] Example prompt sentence:

[0613] """

[0614] Your goal is to write a program that tracks social media sentiment related to a particular brand and analyzes that sentiment. You will use the Twitter API for data collection and natural language processing techniques for data preprocessing and sentiment analysis. You will set up a web server using Flask to display the results in real time.

[0615] """

[0616] Using this system, companies can quickly and accurately grasp market trends and implement effective marketing and sales strategies.

[0617] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0618] Step 1:

[0619] The server selects information sources specified by the user and collects data. Specifically, when the user specifies a specific brand name or related hashtag, the server collects social media posts via the Twitter API. It also accesses an internal database to obtain related purchase history and customer information. The input is the specified brand name and hashtag, and the output is the collected social media posts and customer data.

[0620] Step 2:

[0621] The server preprocesses the collected data and filters out unnecessary information. Specifically, it uses natural language processing techniques to remove spam messages and irrelevant data and standardize the format of the text data. This process involves, for example, using automated scripts to remove URLs and mentions and remove stop words. The input is the collected social media posts and customer data, and the output is the preprocessed data.

[0622] Step 3:

[0623] The preprocessed data is input into a generative AI model to analyze brand awareness, reputation, and target user attributes. The server uses a machine learning model to perform sentiment analysis on the text data and classify the data as positive, negative, or neutral. It also generates a target user profile based on the user's purchasing behavior and hobbies. The input is the preprocessed data, and the output is the sentiment analysis results and the target user profile.

[0624] Step 4:

[0625] The server displays the analysis results in real time on a dashboard and provides them to users. WebSocket technology is used to continuously update the latest analysis results to the dashboard's UI components, allowing users to grasp market trends in real time. The input is the sentiment analysis results and target user profiles, and the output is real-time analysis information displayed on the dashboard.

[0626] Step 5:

[0627] The server uses a generative AI model to predict future trends. Based on time-series data, it predicts fluctuations in popularity and reputation over the next few months and displays the results on a dashboard. Specifically, it uses a predictive algorithm to calculate future data points and visualizes the results. The input is time-series data, and the output is a forecast of future trends.

[0628] Step 6:

[0629] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user. Specifically, it proposes the content and timing of promotions to be carried out in the next campaign. The server sends the generated strategy information to the user via email or push notification. The input is the analysis and prediction results, and the output is a proposed marketing and sales strategy.

[0630] Step 7:

[0631] Users provide feedback to the server on the results of their implemented campaigns and strategies. The server uses this feedback to retrain the AI ​​model and improve the accuracy of the system. For example, a user inputs the effectiveness of a campaign, and the server collects that data and uses it for the next model training. The input is the user's feedback data, and the output is a retrained AI model.

[0632] 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.

[0633] This invention is a system that utilizes generative AI to grasp brand awareness, reputation, and target audience insights in real time, and combines it with an emotion engine that recognizes user emotions to enhance marketing and sales strategies. This system is composed of multiple servers, terminals, and users.

[0634] System Configuration

[0635] 1. Data Collection Methods

[0636] The server accesses external APIs and internal databases to collect data related to a specific brand. For example, the server uses the Twitter API to collect tweets containing the brand name and related hashtags. At the same time, it retrieves past purchase history and customer information from the internal database.

[0637] 2. Data preprocessing methods

[0638] The server preprocesses the collected data and filters out unnecessary information. Specifically, the server removes spam messages and irrelevant data and standardizes the format of the text data. During this process, the server analyzes the text data using natural language processing technology.

[0639] 3. Data Analysis Methods

[0640] The server inputs the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target audience attributes. The server then uses a machine learning model to perform sentiment analysis and classify the data as positive, negative, or neutral. It also generates a target audience profile based on user purchasing behavior and hobbies.

[0641] 4. Real-time feedback methods

[0642] The server displays the analysis results in real time on a dashboard and provides them to the user. For example, the server uses WebSocket technology to continuously update the latest analysis results to the dashboard UI component, allowing users to grasp market trends in real time.

[0643] 5. Means of predicting future trends

[0644] Sarver uses generative AI models to predict future trends for brands. Based on time-series data, it predicts fluctuations in awareness and reputation over the next few months and displays the results on a dashboard. For example, Sarver predicts that brand awareness will increase by 5% next month.

[0645] 6. Strategy Generation Method

[0646] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user. Specifically, the server makes strategy suggestions such as "recommending providing a promotion code in the next campaign." These suggestions are notified to the user not only via the dashboard but also by email and push notifications.

[0647] 7. Feedback channels

[0648] Users provide feedback on the results of their campaigns and strategies to the server. The server uses this feedback to retrain the AI ​​model and improve the accuracy of the system. For example, a user can input the results of a campaign, and the server can collect that data and use it for the next model training.

[0649] 8. Emotional Engine Means

[0650] The server uses an emotion engine to recognize the user's emotions in real time. For example, the server performs emotion analysis based on data acquired from the user's online activities (comments, purchase history, access logs, etc.) to identify the user's current emotional state.

[0651] 9. Personalized Strategy Generation

[0652] The server generates and provides personalized marketing and sales strategies to the user based on the user's emotional state, for example, promoting a specific promotion if the user is in a positive emotional state and providing customer support information if the user is in a negative emotional state.

[0653] Specific examples

[0654] For example, a beverage manufacturer could use this system when launching a new product. The server collects related tweets on the "new beverage name" from social media and analyzes their popularity in real time. The server also combines this with past purchasing data to identify that "sports-loving men in their 30s" are the primary target audience. The generative AI model then predicts that the product's popularity will increase next month. Furthermore, the emotion engine analyzes the emotional state of users from their comments and feedback and provides personalized strategies. For example, if there are a lot of positive reactions, a limited promotional campaign can be promoted, and if there are a lot of negative reactions, immediate customer support can be provided.

[0655] The above is a specific embodiment of the present invention. By using this system, companies can quickly and accurately grasp market trends, implement effective marketing and sales strategies, and even provide personalized responses that take into account the emotional state of users.

[0656] The processing flow will be explained below.

[0657] Step 1: Select a data source

[0658] The server selects a data source to allow it to collect data from an external API (e.g., Twitter API) or an internal database. The server obtains an API key and authentication information to establish access to the external API.

[0659] Step 2: Getting the data

[0660] The server sends a request containing specific keywords or hashtags to retrieve data from selected data sources. Specifically, the server sends an API request to search tweets and posts for the brand name and related keywords.

[0661] Step 3: Preprocessing the data

[0662] The server preprocesses the retrieved data and filters out unnecessary information, specifically filtering out spam messages and normalizing the text (e.g., converting to lowercase and removing extra whitespace). In this step, natural language processing techniques are used to analyze the text data.

[0663] Step 4: Save your data

[0664] The server stores the pre-processed data in an internal database for quick access in subsequent processing steps.

[0665] Step 5: Perform sentiment analysis

[0666] The server uses a generative AI model to perform sentiment analysis on the pre-processed data, for example, taking tweet or review text as input and classifying it as positive, negative, or neutral.

[0667] Step 6: Target Audience Analysis

[0668] The server combines the results of the sentiment analysis with the user's purchasing history to perform a detailed analysis of the target audience's attributes, specifically creating a profile of their age group, gender, hobbies, and preferences.

[0669] Step 7: Provide real-time feedback

[0670] The server displays the analysis results in a dashboard in real time, using WebSocket or API to ensure that data is reflected to users in real time.

[0671] Step 8: Perform future trend predictions

[0672] The server uses a generative AI model to predict future trends for the brand, predicting fluctuations in name recognition and reputation based on time-series data.

[0673] Step 9: Providing prediction results

[0674] The server displays the prediction results on a dashboard and provides them to the user, visualizing them in graphs and charts for easy understanding.

[0675] Step 10: Generate a marketing and sales strategy

[0676] Based on the analysis and prediction results, the server proposes specific marketing and sales strategies to the user, such as providing promotion codes and recommending targeted campaigns.

[0677] Step 11: Communicate your strategy

[0678] The server notifies the user of the generated marketing and sales strategies via the dashboard, email, and push notifications.

[0679] Step 12: Collect user emotion data

[0680] The server collects data on the user's emotional state from their online activities (comments, purchase history, access logs, etc.) and uses an emotion engine to analyze the collected data in real time to identify the user's emotional state.

[0681] Step 13: Individualized strategy generation

[0682] The server generates and provides personalized marketing and sales strategies based on the user's emotional state, for example, promoting new product introductions and promotional campaigns to users in a positive emotional state, and providing customer support information to users in a negative emotional state.

[0683] Step 14: Getting feedback

[0684] Users report the results of their campaigns and strategies to the server, specifically by inputting the success rate and KPIs (key performance indicators) of the campaigns.

[0685] Step 15: Retrain the model

[0686] The server uses feedback data obtained from users to retrain the generative AI model and emotion engine to improve accuracy, and adds new data to continuously improve the model's performance.

[0687] The above are the specific processing steps for carrying out the invention.

[0688] Example 2

[0689] 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."

[0690] Traditional marketing and sales strategies struggle to collect and analyze large amounts of data in real time to generate effective strategies. It's also difficult to provide personalized responses that take into account the user's emotional state. Furthermore, there's a lack of a mechanism for continuously improving models based on feedback, limiting the accuracy of the system.

[0691] 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: means for selecting information sources from users and accessing external APIs and internal storage devices; means for preprocessing the collected information and filtering unnecessary information; means for inputting the preprocessed information into a generative AI model and analyzing brand awareness, reputation, and target audience attributes; means for displaying the analysis results on a display device in real time and providing them to the user; means for predicting future trends and providing the prediction results; means for generating sales promotion and sales tactics based on the analysis and prediction results and notifying the user; means for obtaining user feedback and retraining the model; means for recognizing user emotions in real time using an emotion engine; and means for generating personalized strategies based on the user's emotional state and providing them to the user. This enables rapid and accurate understanding of market trends and individualized responses that take the user's emotional state into consideration. Furthermore, by continuously collecting feedback and improving the model, the accuracy of the system can be improved.

[0692] "User" refers to the entity that uses this system to execute marketing and sales strategies.

[0693] "Source" refers to an external API or internal storage device from which data is obtained.

[0694] "External API" refers to an application program interface for obtaining data from external services provided on the Internet.

[0695] "Internal storage device" refers to a database that stores past purchase history, customer information, etc., managed within a company.

[0696] "Collected Information" refers to data obtained by the Server from sources designated by the User.

[0697] "Preprocessing" refers to the process of filtering unnecessary information from collected information and standardizing the format of text data.

[0698] A "generative AI model" refers to an artificial intelligence model that can analyze data and make predictions in a generative way.

[0699] "Brand awareness" refers to an indicator of how familiar consumers are with a particular brand.

[0700] "Reputation" refers to the general evaluation or opinion of a brand or product.

[0701] A "target audience" refers to a group of customers who may be interested in a particular product or service.

[0702] "Display Device" refers to a device (e.g., dashboard, monitor) for visually displaying collected and analyzed data.

[0703] "Future trends" refers to predicted changes and trends in the market or brand in the future.

[0704] "Sales promotion" refers to the strategies and measures taken to promote a particular product or service in the market.

[0705] "Sales tactics" refer to the specific approaches and methods used to effectively sell a company's products or services.

[0706] "Feedback" refers to information provided to the server regarding the results of campaigns and strategies implemented by the user.

[0707] "Model retraining" refers to the process of improving a generative AI model based on feedback to improve its performance.

[0708] "Emotion engine" refers to a system component for identifying and analyzing user emotions in real time.

[0709] An "individualized strategy" refers to a strategy that provides optimal marketing and sales tactics to specific users based on their emotional state or specific attributes.

[0710] The system uses generative AI to understand brand awareness, reputation, and target audience insights in real time, combined with an emotion engine that recognizes user emotions to enhance marketing and sales strategies. The entire system is comprised of multiple servers, terminals, and users.

[0711] Hardware and Software Configuration

[0712] The server is primarily responsible for data collection, preprocessing, analysis, real-time feedback, prediction, strategy generation, and emotion recognition. The specific hardware used includes a network interface for processing API requests, a database server for storing data, and a high-performance GPU for running the generative AI model. The software includes the Twitter API, an SQL database, natural language processing libraries (e.g., NLTK and Spacy), and machine learning frameworks (e.g., TensorFlow and PyTorch).

[0713] The terminal is a computer or smartphone that users use to access the system and check analysis results and strategic proposals. The terminal connects to the server via a web browser or dedicated application.

[0714] Examples of data collection

[0715] The server collects data related to a specific brand by accessing external APIs (such as the Twitter API) or an internal database. For example, the server collects tweets containing "new drink name" and other related hashtags. At the same time, the server retrieves past purchase history and customer information from the internal database using SQL queries.

[0716] Example prompt sentence:

[0717] Get the latest tweets related to "New Drink Names".

[0718] Specific examples of data preprocessing

[0719] The server preprocesses the collected data, filtering out unnecessary information, removing spam messages and irrelevant data, and standardizing the format of text data. Specifically, it uses natural language processing technologies (e.g., NLTK and Spacy) to clean up and standardize the text format of tweets.

[0720] Example prompt sentence:

[0721] Please remove spam messages from your data and standardize the format.

[0722] Specific examples of data analysis

[0723] The server inputs the preprocessed data into a generative AI model (e.g., BERT or GPT-3) to analyze brand awareness, reputation, and target audience attributes. It also uses a machine learning model to perform sentiment analysis and classify the data as positive, negative, or neutral. It also generates a target audience profile based on user purchasing behavior and hobbies.

[0724] Example prompt sentence:

[0725] Analyze brand sentiment and categorize sentiment based on pre-processed data.

[0726] Examples of real-time feedback

[0727] The server displays the analysis results in real time on a dashboard and provides them to users. This uses WebSocket technology to continuously update the latest analysis results on the dashboard UI component, allowing users to grasp market trends in real time.

[0728] Example prompt sentence:

[0729] View the latest analytics on your dashboard.

[0730] Specific examples of future trend predictions

[0731] The server uses generative AI models to predict future trends for the brand. For example, the server creates a predictive model based on time-series data to forecast fluctuations in brand awareness and reputation over the next few months. The results are displayed on a dashboard.

[0732] Example prompt sentence:

[0733] Predict your brand awareness for the next three months.

[0734] Example of strategy generation

[0735] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user. Specifically, it makes strategic suggestions such as "We recommend offering a promotion code in the next campaign." These suggestions are not only displayed on the dashboard, but also sent to the user via email and push notifications.

[0736] Example prompt sentence:

[0737] Generate recommended strategies for your next campaign.

[0738] Specific examples of feedback

[0739] Users provide feedback on the results of their campaigns and strategies to the server. The server receives this feedback and retrains the AI ​​model to improve the accuracy of the system. For example, users input campaign results through a dashboard or dedicated app, and the server collects the data and uses it for the next model training.

[0740] Example prompt sentence:

[0741] Save the campaign results as feedback.

[0742] Examples of emotion engines

[0743] The server uses an emotion engine to recognize users' emotions in real time. It performs emotion analysis based on the user's online activities (comments, purchase history, access logs, etc.) to identify their current emotional state. For example, it calculates an emotion score based on the user's comments and purchase history and classifies them as positive, negative, or neutral.

[0744] Example prompt sentence:

[0745] Perform sentiment analysis on user comments.

[0746] Examples of personalized strategies

[0747] The server generates and provides personalized marketing and sales strategies based on the user's emotional state, promoting specific promotions in the case of a positive emotional state and providing customer support information in the case of a negative emotional state.

[0748] Example prompt sentence:

[0749] Generate strategies based on the user's emotional state.

[0750] Using this system, companies can quickly and accurately grasp market trends, implement effective marketing and sales strategies, and even provide personalized responses that take into account the user's emotional state.

[0751] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0752] Step 1: Data collection

[0753] The server collects data from information sources. Specifically, the server accesses an external API (e.g., Twitter API) and retrieves tweets containing specified keywords (e.g., "new beverage name"). The server also retrieves past purchase history and customer information from its internal storage device using SQL queries. The input here is the information source and keywords and conditions to be collected, and the output is the collected raw data.

[0754] Specific behavior:

[0755] The server periodically issues requests to the Twitter API and parses the API responses to retrieve data.

[0756] The server extracts the required information from the internal storage using SQL queries.

[0757] Step 2: Data Preprocessing

[0758] The server preprocesses the collected data and filters out unnecessary information. Specifically, the server uses natural language processing techniques (e.g., NLTK or Spacy) to clean up the tweet text data and standardize the format. The input here is the collected raw data, and the output is the preprocessed clean data.

[0759] Specific behavior:

[0760] The server cleans the text data, removing spam messages and irrelevant data.

[0761] The server uses regular expressions to remove certain patterns and converts the text data into a unified format.

[0762] Step 3: Data analysis

[0763] The server then inputs the preprocessed data into a generative AI model (e.g., BERT or GPT-3) to analyze brand awareness, reputation, and target audience attributes. The input is the preprocessed clean data, and the output is the analysis results. The server also uses a machine learning model to perform sentiment analysis and classify the data as positive, negative, or neutral.

[0764] Specific behavior:

[0765] The server inputs the preprocessed text into a generative AI model to obtain emotion scores and attribute information.

[0766] The server uses a sentiment analysis model to classify tweets and assign them to positive, negative, or neutral sentiment categories.

[0767] Step 4: Real-time feedback

[0768] The server displays the analysis results in real time on a display device and provides them to the user. The input here is the results of the data analysis, and the output is dashboard data that is updated in real time.

[0769] Specific behavior:

[0770] The server pushes data to the dashboard over a WebSocket connection.

[0771] The dashboard UI components receive the latest data and present it visually to the user.

[0772] Step 5: Predict future trends

[0773] The server uses a generative AI model to predict future trends for the brand, where the input is past data and current analysis results, and the output is predicted future trend data.

[0774] Specific behavior:

[0775] The server predicts the data using a time series prediction model (e.g., ARIMA or LSTM).

[0776] The server sends the prediction results to a dashboard for display to the user.

[0777] Step 6: Strategy Generation

[0778] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user, where the inputs are the analysis and prediction results and the output is the generated strategy proposals.

[0779] Specific behavior:

[0780] The server generates a strategy template based on the analysis and prediction results.

[0781] The server sends the generated strategy to the user via email or push notification.

[0782] Step 7: Gather feedback

[0783] The results of the campaigns and strategies executed by the user are provided as feedback to the server. The input here is the results of the campaigns and strategies, and the output is the collected feedback data. The server uses this feedback to retrain the AI ​​model.

[0784] Specific behavior:

[0785] Users enter campaign results through a dashboard or dedicated app.

[0786] The server collects the feedback data and uses it for the next model training.

[0787] Step 8: Emotion Recognition

[0788] The server uses an emotion engine to recognize the user's emotions in real time, where the input is the user's online activity data and the output is the recognized emotional state.

[0789] Specific behavior:

[0790] The server analyzes the user's comments and purchase history and calculates an emotion score.

[0791] The server appropriately tags and stores the emotional state.

[0792] Step 9: Personalized strategy generation

[0793] The server generates personalized marketing and sales strategies based on the user's emotional state and provides them to the user, where the input is the user's emotional state and the output is personalized strategy proposals.

[0794] Specific behavior:

[0795] The server classifies the user's emotional state and selects an appropriate strategy template.

[0796] The server generates a personalized strategy and notifies the user via email or push notification.

[0797] (Application example 2)

[0798] 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."

[0799] Product quality control is a critical issue in modern factory operations. Mass-produced products, in particular, face a high risk of quality variation, making quality control time-consuming and costly. Traditional quality control systems struggle to identify problems in real time and propose rapid solutions. Systems that can predict future quality trends and propose appropriate improvement measures are also lacking. This creates a demand for advanced systems that can maintain high levels of product quality while improving factory efficiency.

[0800] 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.

[0801] In this invention, the server includes: means for collecting product quality data from sensors and camera devices installed in the factory; means for preprocessing the collected product quality data and filtering incomplete or abnormal data; means for inputting the preprocessed quality data into a generative AI model to monitor, evaluate, and identify problems in real time; means for displaying analysis results in real time on a dashboard for factory workers and immediately notifying them if a problem occurs; means for predicting future quality trends based on past quality data and providing the prediction results to workers; and means for proposing optimal product inspection methods and manufacturing processes based on the analysis and prediction results. This enables highly accurate real-time monitoring and rapid response, not only enabling efficient product quality control but also enabling proactive improvement measures to be implemented based on future trend predictions.

[0802] A "data source" is a source of information from which information is obtained, such as an external API or an internal database.

[0803] "External API" refers to an interface to external services or data provided over the Internet.

[0804] An "internal database" is a system that stores data managed within an organization.

[0805] "Preprocessing" refers to the process of converting raw data into a form that is easier to analyze, such as removing noise and standardizing the format.

[0806] A "generative AI model" is a model that uses generative adversarial networks (GANs) or other AI algorithms to analyze data and generate new information.

[0807] A "brand" is a collective term for the elements that create a particular image and reputation for a company or product.

[0808] "Awareness" is a measure of how many people know about a brand or product.

[0809] "Reputation" refers to the evaluation and feelings that consumers and the market have about a brand or product.

[0810] "Target audience" refers to a specific group of consumers that you are targeting with marketing and advertising.

[0811] "Attributes" refer to characteristics of the target audience, such as gender, age, and hobbies.

[0812] "Real-time" refers to the state in which data is acquired and processed almost immediately.

[0813] A "dashboard" is an interface that visually displays data analysis results and important information.

[0814] "Future trends" refers to future tendencies and developments predicted by analyzing past and present data.

[0815] "Feedback" refers to information used to adjust future actions based on the results and reactions of implemented measures and strategies.

[0816] "Factory" refers to a place where products are produced or processed.

[0817] A "sensor" is a device that senses physical conditions or data and converts them into electrical signals.

[0818] A "camera device" is a device that captures video and converts it into digital data.

[0819] "Quality Data" refers to data relating to the characteristics and performance of a product.

[0820] "Incomplete data" refers to data that is missing or missing.

[0821] "Abnormal data" refers to data that exhibits unusual values ​​or patterns that fall outside of normal ranges.

[0822] "Monitoring" means constantly checking a specific object or condition.

[0823] "Evaluation" means judging the quality or performance of something.

[0824] "Notification" refers to the immediate sharing of specific information or results.

[0825] "Quality Trends" refers to trends or fluctuations in product quality.

[0826] "Improvement measures" refer to methods and means for solving current problems and improving the situation.

[0827] "Product inspection methods" refer to specific methods and techniques for checking product quality.

[0828] "Manufacturing process" refers to the series of steps taken to create a product from raw materials.

[0829] This invention provides a system for streamlining factory product quality control and continuously providing high-quality products. This system utilizes various sensors and camera devices installed in the factory to collect and analyze product quality data in real time. It also proposes quick countermeasures based on the analysis results and predicts future quality trends, thereby improving product quality.

[0830] System Configuration

[0831] The server operates in the following specific steps:

[0832] Data collection methods

[0833] The server collects product quality data using sensors and camera devices installed in the factory, including various quality indicators such as product dimensions, weight, and appearance. Specifically, it uses IoT devices and communication modules to acquire data from the various sensors and cameras via a network.

[0834] Data preprocessing measures

[0835] The server preprocesses the collected data, filtering out incomplete or abnormal data, and utilizes open-source natural language processing libraries (e.g., NLTK) and image processing libraries (e.g., OpenCV) to remove noise from the data and convert it into a unified format.

[0836] Data Analysis Methods

[0837] The preprocessed data is input into a generative AI model. The server uses the generative AI model to monitor product quality in real time and identify problems. A sentiment engine (e.g., Sentiment Analyzer) is also used for analysis, analyzing quality data and feedback logs from workers to identify the cause of quality issues.

[0838] Real-time feedback tools

[0839] The server displays the analysis results in real time on a dashboard for factory workers and immediately notifies them if any problems occur. WebSocket technology is used to continuously update the dashboard's UI components with the latest information, allowing workers to immediately detect problems and take action.

[0840] A means of predicting future trends

[0841] The server predicts future quality trends based on past quality data. To do this, it uses a time series analysis algorithm to predict future fluctuations in product quality. The analysis results are displayed on a dashboard and provided to workers.

[0842] Strategy Generation Method

[0843] Based on the analysis and prediction results, the server proposes optimal product inspection methods and manufacturing processes, allowing factory workers to quickly implement appropriate improvement measures.

[0844] Specific examples

[0845] For example, this system is used in a factory to control the quality of steel products. The server collects data on product dimensions, weight, and surface condition from sensors and camera devices, and evaluates the quality in real time. The preprocessed data is input into a generative AI model to detect quality anomalies. The detection results are displayed on a dashboard and workers are immediately notified. Furthermore, future quality trends are predicted based on past data, and if a prediction is made that the quality of steel products will improve with a certain probability next month, specific improvement measures are suggested to workers.

[0846] Prompt Sentence Examples

[0847] Input product quality data and previous day's manufacturing process data into a generative AI model and suggest your next quality control strategy. For example, are there any trends like the following?

[0848] By using the system of the present invention, not only can factory quality control be carried out efficiently, but proactive improvement measures can also be implemented based on predictions of future trends.

[0849] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0850] Step 1: Data collection

[0851] The server collects product quality data from sensors and camera devices installed in the factory. The sensors acquire physical data such as product dimensions, weight, and temperature, while the camera devices acquire images of the product's appearance. The input is sensor and camera data, and the output is the collected results of this data. Specifically, the server accesses the API of each device, requests data, and acquires it.

[0852] Step 2: Data Preprocessing

[0853] The server preprocesses the collected data and filters out incomplete or abnormal data. Specifically, it performs missing value imputation, outlier detection and removal, and data format standardization. The input is the data collected in step 1, and the output is quality data that has been filtered and cleaned. Specifically, it performs these processes using a natural language processing library (NLTK) and an image processing library (OpenCV).

[0854] Step 3: Data analysis

[0855] The server inputs the preprocessed data into the generative AI model to monitor, evaluate, and identify problems in real time with product quality. The input is preprocessed quality data, and the output is the quality evaluation results. Specifically, the server inputs data into the generative AI model and uses a sentiment engine (Sentiment Analyzer) to evaluate the quality.

[0856] Step 4: Real-time feedback

[0857] The server displays the analysis results in real time on a dashboard for factory workers and immediately notifies them if any problems occur. The dashboard is updated using WebSocket technology. The input is the quality evaluation results, and the output is real-time notifications to workers and the dashboard display. Specifically, the UI components are updated via the WebSocket connection.

[0858] Step 5: Predict future trends

[0859] The server predicts future quality trends based on past quality data. It uses a time series analysis algorithm to predict future quality fluctuations. The input is past and current quality data, and the output is the future quality trend prediction result. Specifically, it performs predictive analysis using a time series analysis algorithm.

[0860] Step 6: Strategy Generation

[0861] The server proposes optimal product inspection methods and manufacturing processes based on the analysis and prediction results. The inputs are prediction results and quality assessment results, and the output is improvement proposals. Specifically, the server notifies the dashboard and workers of the optimal strategy derived by the generative AI model.

[0862] Step 7: Get feedback and retrain

[0863] The server receives feedback on the implemented improvements and retrains the generative AI model. The input is feedback data from the workers, and the output is the retrained generative AI model. Specifically, the server uses the received feedback data to update the AI ​​model's parameters and improve its accuracy.

[0864] 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.

[0865] 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.

[0866] 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.

[0867] [Third embodiment]

[0868] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0869] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0870] 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).

[0871] 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.

[0872] 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.

[0873] 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).

[0874] 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.

[0875] 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.

[0876] 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.

[0877] 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.

[0878] 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.

[0879] 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."

[0880] This invention is a system that utilizes generative AI to understand brand awareness, reputation, and target audience insights in real time to enhance marketing and sales strategies. This system is composed of multiple servers, terminals, and users.

[0881] System Configuration

[0882] 1. Data Collection Methods

[0883] The server accesses external APIs and internal databases to collect data related to a specific brand. For example, the server uses the Twitter API to collect tweets containing the brand name and related hashtags. At the same time, it retrieves past purchase history and customer information from the internal database.

[0884] 2. Data preprocessing methods

[0885] The server preprocesses the collected data and filters out unnecessary information. Specifically, the server removes spam messages and irrelevant data and standardizes the format of the text data. During this process, the server analyzes the text data using natural language processing technology.

[0886] 3. Data Analysis Methods

[0887] The server inputs the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target audience attributes. At this stage, the server uses a machine learning model to perform sentiment analysis and classify the data as positive, negative, or neutral. It also generates a target audience profile based on user purchasing behavior and hobbies.

[0888] 4. Real-time feedback methods

[0889] The server displays the analysis results in real time on a dashboard and provides them to the user. For example, the server uses WebSocket technology to continuously update the latest analysis results to the dashboard UI component, allowing users to grasp market trends in real time.

[0890] 5. Means of predicting future trends

[0891] Sarver uses generative AI models to predict future trends for brands. Based on time-series data, it predicts fluctuations in awareness and reputation over the next few months and displays the results on a dashboard. For example, Sarver predicts that brand awareness will increase by 5% next month.

[0892] 6. Strategy Generation Method

[0893] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user. Specifically, the server makes strategy suggestions such as "recommending providing a promotion code in the next campaign." These suggestions are notified to the user not only via the dashboard but also by email and push notifications.

[0894] 7. Feedback channels

[0895] Users provide feedback on the results of their campaigns and strategies to the server. The server uses this feedback to retrain the AI ​​model and improve the accuracy of the system. For example, a user can input the results of a campaign, and the server can collect that data and use it for the next model training.

[0896] Specific examples

[0897] For example, a beverage manufacturer might use this system when launching a new product. The server collects tweets related to the "new beverage name" from social media and analyzes its popularity in real time. The server also combines this with past purchasing data to identify that "men in their 30s who love sports" are the primary target audience. The generative AI model then predicts that the product's popularity will increase next month. Based on this information, the system suggests a strategy to the user, such as "running a promotion during a sporting event," and provides feedback on the results to be used next time.

[0898] The above is a specific embodiment of the present invention. By using this system, companies can quickly and accurately grasp market trends and implement effective marketing and sales strategies.

[0899] The processing flow will be explained below.

[0900] Step 1: Select a data source

[0901] A data source is selected so that the server can collect data from an external API (e.g., Twitter API) or an internal database. The server obtains an API key and authentication information to establish access to the external API.

[0902] Step 2: Getting the data

[0903] The server sends a request containing specific keywords or hashtags to retrieve data from selected data sources. Specifically, the server sends an API request to search tweets and posts for the brand name and related keywords.

[0904] Step 3: Preprocessing the data

[0905] The server preprocesses the retrieved data and filters out unnecessary information, specifically filtering out spam messages and normalizing the text (e.g., converting to lowercase and removing extra whitespace). In this step, natural language processing techniques are used to analyze the text data.

[0906] Step 4: Save your data

[0907] The server stores the pre-processed data in an internal database for quick access in subsequent processing steps.

[0908] Step 5: Perform sentiment analysis

[0909] The server uses a generative AI model to perform sentiment analysis on the pre-processed data, for example, taking tweet or review text as input and classifying it as positive, negative, or neutral.

[0910] Step 6: Target Audience Analysis

[0911] The server combines the results of the sentiment analysis with the user's purchasing history to perform a detailed analysis of the target audience's attributes, specifically creating a profile of their age group, gender, hobbies, and preferences.

[0912] Step 7: Provide real-time feedback

[0913] The server displays the analysis results in a dashboard in real time, using WebSocket or API to ensure that data is reflected to users in real time.

[0914] Step 8: Perform future trend predictions

[0915] The server uses a generative AI model to predict future trends for the brand, predicting fluctuations in name recognition and reputation based on time-series data.

[0916] Step 9: Providing prediction results

[0917] The server displays the prediction results on a dashboard and provides them to the user, visualizing them in graphs and charts for easy understanding.

[0918] Step 10: Generate a marketing and sales strategy

[0919] Based on the analysis and prediction results, the server proposes specific marketing and sales strategies to the user, such as providing promotion codes and recommending targeted campaigns.

[0920] Step 11: Communicate your strategy

[0921] The server notifies the user of the generated marketing and sales strategies via the dashboard, email, and push notifications.

[0922] Step 12: Get feedback

[0923] Users report the results of their campaigns and strategies to the server, specifically by inputting the success rate and KPIs (key performance indicators) of the campaigns.

[0924] Step 13: Retrain the model

[0925] The server uses feedback data obtained from users to retrain the generative AI model to improve its accuracy, and adds new data to continuously improve the model's performance.

[0926] The above are the specific processing steps for carrying out the invention.

[0927] Example 1

[0928] 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."

[0929] Traditional methods for formulating marketing and sales strategies make it difficult to grasp real-time market trends, making it difficult to quickly derive effective strategies. Furthermore, there is a lack of a process for improving the system based on feedback, making it difficult to improve accuracy. Furthermore, existing systems require a great deal of time and effort to preprocess and analyze massive amounts of data, making efficient operation difficult.

[0930] 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.

[0931] In this invention, the server includes means for selecting a data source from a user and accessing an external API or an internal database, means for preprocessing the collected data and filtering out unnecessary information, means for inputting the preprocessed data into a generative AI model and analyzing brand awareness, reputation, and target audience attributes, means for displaying the analysis results on a dashboard in real time and providing them to the user, means for predicting future trends and providing the prediction results, means for generating a marketing and sales strategy using the generative AI model and notifying the user, and means for obtaining user feedback and retraining the model. This makes it possible to grasp market trends in real time and formulate highly accurate marketing and sales strategies.

[0932] A "data source" is a reference from which data is collected, such as an external API or an internal database.

[0933] "Preprocessing" is the process of filtering unnecessary information from collected data and standardizing the data format.

[0934] A "generative AI model" is a model that uses machine learning algorithms to analyze data and generate insights.

[0935] "Brand awareness" is an indicator of how well a particular brand is known among the general public.

[0936] "Reputation" refers to consumer opinions and ratings of a particular brand or product.

[0937] A "target audience" is a group of consumers that is the target of marketing and advertising.

[0938] "Displaying on the dashboard in real time" means that the analysis results are immediately reflected in the user interface so that the user can check them immediately.

[0939] "Predicting future trends" means predicting future market trends and fluctuations in brand awareness based on collected data.

[0940] A "marketing and sales strategy" is a strategic plan for increasing sales of a product or service.

[0941] "User notification" is the process of informing users of generated strategies and important information.

[0942] "Feedback" is any opinion or evaluation provided based on the results of an implemented strategy or campaign.

[0943] "Model retraining" is the re-learning process of using collected feedback data to improve the accuracy of an existing generative AI model.

[0944] This invention is a system that utilizes generative AI to understand brand awareness, reputation, and target audience insights in real time to enhance marketing and sales strategies. This system is composed of multiple servers, terminals, and users.

[0945] System Configuration

[0946] 1. Data Collection Methods

[0947] The server accesses external APIs and internal databases to collect data related to a specific brand. For example, the server uses the Twitter API to collect tweets containing a specific brand name or related hashtags. The server also retrieves past purchase history and customer information from internal databases. This data is obtained from databases such as MongoDB and MySQL.

[0948] 2. Data preprocessing methods

[0949] The server preprocesses the collected data and filters out unnecessary information. Specifically, the server removes spam messages and irrelevant data and standardizes the format of the text data. During this process, the server uses natural language processing techniques to analyze the text data using the Python libraries NLTK and spaCy.

[0950] 3. Data Analysis Methods

[0951] The server inputs the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target audience attributes. At this stage, the server performs sentiment analysis using the Hugging Face Transformer model to classify the data as positive, negative, or neutral. The server also uses a clustering algorithm (e.g., K-means) to generate a target audience profile based on user purchasing behavior and hobbies.

[0952] 4. Real-time feedback methods

[0953] The server displays the analysis results in real time on a dashboard and provides them to the user. For example, the server uses WebSocket technology to continuously update the dashboard's UI components with the latest analysis results. The front-end terminal uses React.js and D3.js to update graphs and charts based on the received data.

[0954] 5. Means of predicting future trends

[0955] The server uses a generative AI model to predict future trends for the brand. Based on time-series data, it uses a Long Short-Term Memory (LSTM) network to predict fluctuations in awareness and reputation over the next few months, and displays the results on a dashboard.

[0956] 6. Strategy Generation Method

[0957] The server uses the generative AI model to generate marketing and sales strategies and notify users. Specifically, the server uses the generative AI model to propose strategies such as "recommend offering a promotional code in the next campaign" based on demand forecasts. These strategies are not only displayed on the dashboard, but also notified to users via email and push notifications using AWS SNS (Simple Notification Service).

[0958] 7. Feedback channels

[0959] Users provide feedback to the server on the results of their campaigns and strategies. They input the results data using their devices and send it to the server. The server then retrains the generative AI model based on the collected feedback data, improving the accuracy of the next analysis and prediction. TensorFlow and PyTorch are used for retraining.

[0960] Specific examples

[0961] For example, a beverage manufacturer might use this system when launching a new product. The server collects tweets related to the "new beverage name" from social media and analyzes its popularity in real time. The server also combines this with past purchasing data to identify that "men in their 30s who love sports" are the primary target audience. The generative AI model then predicts that the product's popularity will increase next month. Based on this information, the system suggests a strategy to the user, such as "running a promotion during a sporting event," and provides feedback on the results to be used next time.

[0962] (Example of a prompt)

[0963] Collect data to analyze your brand awareness and reputation.

[0964] Preprocess the collected data to filter out irrelevant data.

[0965] Analyze the preprocessed data and classify it as positive, negative, or neutral.

[0966] Predict your brand's future reputation and visibility.

[0967] Based on the analysis, suggest a strategy for your next campaign.

[0968] The above is a specific embodiment of the present invention. By using this system, companies can quickly and accurately grasp market trends and implement effective marketing and sales strategies.

[0969] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0970] Step 1: Data source selection and data collection

[0971] 1. Input: The user selects a data source containing brand names and related keywords.

[0972] 2. Processing: The server accesses external APIs and internal databases based on the input information received from the user. Specifically, it uses the Twitter API to collect tweets containing the specified brand name or hashtag. It also retrieves past purchase history and customer information from internal databases (e.g., MongoDB or MySQL).

[0973] 3. Output: The collected data is stored in the server and is ready for the next step.

[0974] Step 2: Data Preprocessing

[0975] 1. Input: Data collected in step 1.

[0976] 2. Processing: The server preprocesses the data, removing spam messages and irrelevant data and standardizing the format of the text data. Specifically, it uses the Python libraries NLTK and spaCy to parse the text data and convert the format.

[0977] 3. Output: Preprocessed and clean data is generated and sent to the next step.

[0978] Step 3: Input and analysis into the generative AI model

[0979] 1. Input: The preprocessed data from step 2.

[0980] 2. Processing: The server inputs this data into a generative AI model. It uses Hugging Face's Transformer model to perform sentiment analysis and classify the data as positive, negative, or neutral. It also uses a clustering algorithm (e.g., K-means) to generate a profile of the target audience.

[0981] 3. Output: The analyzed data is generated and results are obtained regarding awareness, reputation, and target audience attributes.

[0982] Step 4: Real-time feedback

[0983] 1. Input: The analysis results generated in step 3.

[0984] 2. Processing: The server reflects the results in real time on the dashboard. Using WebSocket technology, the latest analysis results are updated sequentially on the dashboard's UI components. The terminal visualizes the data using React.js and D3.js.

[0985] 3. Output: The latest data is displayed in the dashboard that users access.

[0986] Step 5: Predict future trends

[0987] 1. Input: Analysis results and time series data generated in Step 3.

[0988] 2. Processing: The server uses a Long Short-Term Memory (LSTM) network to predict future fluctuations in popularity and reputation.

[0989] 3. Output: The prediction results are generated and displayed on a dashboard.

[0990] Step 6: Strategy Generation

[0991] 1. Input: The analytical and prediction results generated in Step 3 and Step 5.

[0992] 2. Processing: The server uses the generative AI model to generate marketing and sales strategies, such as providing promotional codes for the next campaign.

[0993] 3. Output: The generated strategies are displayed on the dashboard and notified to the user via email and push notifications using AWS SNS (Simple Notification Service).

[0994] Step 7: Get feedback and retrain the model

[0995] 1. Input: The user inputs the results of the campaigns and strategies they have executed using a terminal and sends them to the server.

[0996] 2. Processing: The server uses the collected feedback data to retrain the generative AI model using TensorFlow or PyTorch.

[0997] 3. Output: A retrained AI model is generated, improving the accuracy of the next analysis or prediction.

[0998] (Application example 1)

[0999] 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."

[1000] It has been difficult to accurately understand the effectiveness of traditional advertising campaigns in real time and adjust strategies immediately. It has also been difficult to generate detailed profiles of target users based on their purchasing behavior and hobbies. Furthermore, there has been a lack of efficient means to process large amounts of data obtained from social media and provide sentiment analysis results in real time.

[1001] 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.

[1002] In this invention, the server includes means for selecting information sources from users and accessing external APIs and internal databases, means for preprocessing the collected data and filtering unnecessary information, means for inputting the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target user attributes, means for displaying sentiment analysis results collected from social media on a dashboard in real time and providing them to users, means for predicting future trends and providing the prediction results, means for generating marketing and sales strategies based on the analysis results and prediction results and notifying users, and means for obtaining user feedback and retraining the model. This makes it possible to accurately grasp the effectiveness of advertising campaigns in real time and immediately adjust strategies.

[1003] "User" means any person or entity that uses this system.

[1004] A "source" is a service or system that provides data, such as an external API or an internal database.

[1005] A "generative AI model" is an artificial intelligence model that is trained to take data as input and analyze brand awareness, reputation, and target user attributes.

[1006] "Social media" refers to online platforms such as Twitter where users post and share information.

[1007] "Sentiment analysis" is the process of classifying the sentiment of text data into positive, negative, or neutral.

[1008] A "dashboard" is an interface for displaying analysis results and prediction results in real time.

[1009] A "marketing strategy" is a plan or method aimed at promoting a product or service.

[1010] A "sales strategy" refers to the plans and methods of sales activities aimed at increasing sales.

[1011] "Feedback" refers to information received from users regarding the results of implemented campaigns and strategies.

[1012] "Retraining" is the process of retraining an artificial intelligence model based on feedback data.

[1013] "Future trends" refers to future trends predicted based on time series data and generative AI models.

[1014] This invention is a system that utilizes generative AI models to understand brand awareness, reputation, and target user insights in real time to enhance marketing and sales strategies. The system consists of a server and an application that can be accessed from a smartphone or other device. A specific example is shown below.

[1015] Functional Configuration

[1016] 1. Data Collection Methods

[1017] The server selects the information source specified by the user and accesses external APIs (e.g., Twitter API) or internal databases. The server collects data containing the specified brand name and related hashtags. This data includes social media posts, the user's past purchase history, customer information, etc.

[1018] 2. Data preprocessing methods

[1019] The server preprocesses the collected data and filters out unnecessary information, using natural language processing techniques to remove spam messages and irrelevant data and standardize the format of the text data.

[1020] 3. Data Analysis Methods

[1021] The preprocessed data is input into a generative AI model, and the server analyzes brand awareness, reputation, and target user attributes. A machine learning model is used to perform sentiment analysis and classify the data as positive, negative, or neutral. It also generates target user profiles based on user purchasing behavior and hobbies.

[1022] 4. Real-time feedback methods

[1023] The server displays the analysis results in real time on a dashboard and provides them to users. WebSocket technology is used to continuously update the dashboard UI components with the latest analysis results, allowing users to grasp market trends in real time.

[1024] 5. Means of predicting future trends

[1025] The server uses generative AI models to predict future trends, using time-series data to forecast fluctuations in popularity and reputation over the next few months, and displays the results on a dashboard.

[1026] 6. Strategy Generation Method

[1027] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user of them, specifically proposing the content and timing of promotions to be carried out in the next campaign.

[1028] 7. Feedback channels

[1029] Users provide feedback on the results of their campaigns and strategies to the server, which then uses this feedback to retrain the AI ​​model and improve the system's accuracy, enabling it to provide more effective strategies for the next campaign.

[1030] Examples of specific examples and prompts

[1031] For example, a beverage manufacturer might use this system when running an advertising campaign for a new product. The server collects tweets related to the "new beverage name" from social media and analyzes their popularity in real time. The server also combines this with past purchasing data to identify that "men in their 30s who love sports" are the primary target users. Based on this information, the generative AI model predicts that "product popularity will increase next month" and suggests a strategy to users: "run a promotion during a sporting event." When the results of the campaign are fed back, the AI ​​model is retrained using that data.

[1032] Example prompt sentence:

[1033] """

[1034] Your goal is to write a program that tracks social media sentiment related to a particular brand and analyzes that sentiment. You will use the Twitter API for data collection and natural language processing techniques for data preprocessing and sentiment analysis. You will set up a web server using Flask to display the results in real time.

[1035] """

[1036] Using this system, companies can quickly and accurately grasp market trends and implement effective marketing and sales strategies.

[1037] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1038] Step 1:

[1039] The server selects information sources specified by the user and collects data. Specifically, when the user specifies a specific brand name or related hashtag, the server collects social media posts via the Twitter API. It also accesses an internal database to obtain related purchase history and customer information. The input is the specified brand name and hashtag, and the output is the collected social media posts and customer data.

[1040] Step 2:

[1041] The server preprocesses the collected data and filters out unnecessary information. Specifically, it uses natural language processing techniques to remove spam messages and irrelevant data and standardize the format of the text data. This process involves, for example, using automated scripts to remove URLs and mentions and remove stop words. The input is the collected social media posts and customer data, and the output is the preprocessed data.

[1042] Step 3:

[1043] The preprocessed data is input into a generative AI model to analyze brand awareness, reputation, and target user attributes. The server uses a machine learning model to perform sentiment analysis on the text data and classify the data as positive, negative, or neutral. It also generates a target user profile based on the user's purchasing behavior and hobbies. The input is the preprocessed data, and the output is the sentiment analysis results and the target user profile.

[1044] Step 4:

[1045] The server displays the analysis results in real time on a dashboard and provides them to users. WebSocket technology is used to continuously update the latest analysis results to the dashboard's UI components, allowing users to grasp market trends in real time. The input is the sentiment analysis results and target user profiles, and the output is real-time analysis information displayed on the dashboard.

[1046] Step 5:

[1047] The server uses a generative AI model to predict future trends. Based on time-series data, it predicts fluctuations in popularity and reputation over the next few months and displays the results on a dashboard. Specifically, it uses a predictive algorithm to calculate future data points and visualizes the results. The input is time-series data, and the output is a forecast of future trends.

[1048] Step 6:

[1049] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user. Specifically, it proposes the content and timing of promotions to be carried out in the next campaign. The server sends the generated strategy information to the user via email or push notification. The input is the analysis and prediction results, and the output is a proposed marketing and sales strategy.

[1050] Step 7:

[1051] Users provide feedback to the server on the results of their implemented campaigns and strategies. The server uses this feedback to retrain the AI ​​model and improve the accuracy of the system. For example, a user inputs the effectiveness of a campaign, and the server collects that data and uses it for the next model training. The input is the user's feedback data, and the output is a retrained AI model.

[1052] 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.

[1053] This invention is a system that utilizes generative AI to grasp brand awareness, reputation, and target audience insights in real time, and combines it with an emotion engine that recognizes user emotions to enhance marketing and sales strategies. This system is composed of multiple servers, terminals, and users.

[1054] System Configuration

[1055] 1. Data Collection Methods

[1056] The server accesses external APIs and internal databases to collect data related to a specific brand. For example, the server uses the Twitter API to collect tweets containing the brand name and related hashtags. At the same time, it retrieves past purchase history and customer information from the internal database.

[1057] 2. Data preprocessing methods

[1058] The server preprocesses the collected data and filters out unnecessary information. Specifically, the server removes spam messages and irrelevant data and standardizes the format of the text data. During this process, the server analyzes the text data using natural language processing technology.

[1059] 3. Data Analysis Methods

[1060] The server inputs the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target audience attributes. The server then uses a machine learning model to perform sentiment analysis and classify the data as positive, negative, or neutral. It also generates a target audience profile based on user purchasing behavior and hobbies.

[1061] 4. Real-time feedback methods

[1062] The server displays the analysis results in real time on a dashboard and provides them to the user. For example, the server uses WebSocket technology to continuously update the latest analysis results to the dashboard UI component, allowing users to grasp market trends in real time.

[1063] 5. Means of predicting future trends

[1064] Sarver uses generative AI models to predict future trends for brands. Based on time-series data, it predicts fluctuations in awareness and reputation over the next few months and displays the results on a dashboard. For example, Sarver predicts that brand awareness will increase by 5% next month.

[1065] 6. Strategy Generation Method

[1066] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user. Specifically, the server makes strategy suggestions such as "recommending providing a promotion code in the next campaign." These suggestions are notified to the user not only via the dashboard but also by email and push notifications.

[1067] 7. Feedback channels

[1068] Users provide feedback on the results of their campaigns and strategies to the server. The server uses this feedback to retrain the AI ​​model and improve the accuracy of the system. For example, a user can input the results of a campaign, and the server can collect that data and use it for the next model training.

[1069] 8. Emotional Engine Means

[1070] The server uses an emotion engine to recognize the user's emotions in real time. For example, the server performs emotion analysis based on data acquired from the user's online activities (comments, purchase history, access logs, etc.) to identify the user's current emotional state.

[1071] 9. Personalized Strategy Generation

[1072] The server generates and provides personalized marketing and sales strategies to the user based on the user's emotional state, for example, promoting a specific promotion if the user is in a positive emotional state and providing customer support information if the user is in a negative emotional state.

[1073] Specific examples

[1074] For example, a beverage manufacturer could use this system when launching a new product. The server collects related tweets on the "new beverage name" from social media and analyzes their popularity in real time. The server also combines this with past purchasing data to identify that "sports-loving men in their 30s" are the primary target audience. The generative AI model then predicts that the product's popularity will increase next month. Furthermore, the emotion engine analyzes the emotional state of users from their comments and feedback and provides personalized strategies. For example, if there are a lot of positive reactions, a limited promotional campaign can be promoted, and if there are a lot of negative reactions, immediate customer support can be provided.

[1075] The above is a specific embodiment of the present invention. By using this system, companies can quickly and accurately grasp market trends, implement effective marketing and sales strategies, and even provide personalized responses that take into account the emotional state of users.

[1076] The processing flow will be explained below.

[1077] Step 1: Select a data source

[1078] The server selects a data source to allow it to collect data from an external API (e.g., Twitter API) or an internal database. The server obtains an API key and authentication information to establish access to the external API.

[1079] Step 2: Getting the data

[1080] The server sends a request containing specific keywords or hashtags to retrieve data from selected data sources. Specifically, the server sends an API request to search tweets and posts for the brand name and related keywords.

[1081] Step 3: Preprocessing the data

[1082] The server preprocesses the retrieved data and filters out unnecessary information, specifically filtering out spam messages and normalizing the text (e.g., converting to lowercase and removing extra whitespace). In this step, natural language processing techniques are used to analyze the text data.

[1083] Step 4: Save your data

[1084] The server stores the pre-processed data in an internal database for quick access in subsequent processing steps.

[1085] Step 5: Perform sentiment analysis

[1086] The server uses a generative AI model to perform sentiment analysis on the pre-processed data, for example, taking tweet or review text as input and classifying it as positive, negative, or neutral.

[1087] Step 6: Target Audience Analysis

[1088] The server combines the results of the sentiment analysis with the user's purchasing history to perform a detailed analysis of the target audience's attributes, specifically creating a profile of their age group, gender, hobbies, and preferences.

[1089] Step 7: Provide real-time feedback

[1090] The server displays the analysis results in a dashboard in real time, using WebSocket or API to ensure that data is reflected to users in real time.

[1091] Step 8: Perform future trend predictions

[1092] The server uses a generative AI model to predict future trends for the brand, predicting fluctuations in name recognition and reputation based on time-series data.

[1093] Step 9: Providing prediction results

[1094] The server displays the prediction results on a dashboard and provides them to the user, visualizing them in graphs and charts for easy understanding.

[1095] Step 10: Generate a marketing and sales strategy

[1096] Based on the analysis and prediction results, the server proposes specific marketing and sales strategies to the user, such as providing promotion codes and recommending targeted campaigns.

[1097] Step 11: Communicate your strategy

[1098] The server notifies the user of the generated marketing and sales strategies via the dashboard, email, and push notifications.

[1099] Step 12: Collect user emotion data

[1100] The server collects data on the user's emotional state from their online activities (comments, purchase history, access logs, etc.) and uses an emotion engine to analyze the collected data in real time to identify the user's emotional state.

[1101] Step 13: Individualized strategy generation

[1102] The server generates and provides personalized marketing and sales strategies based on the user's emotional state, for example, promoting new product introductions and promotional campaigns to users in a positive emotional state, and providing customer support information to users in a negative emotional state.

[1103] Step 14: Getting feedback

[1104] Users report the results of their campaigns and strategies to the server, specifically by inputting the success rate and KPIs (key performance indicators) of the campaigns.

[1105] Step 15: Retrain the model

[1106] The server uses feedback data obtained from users to retrain the generative AI model and emotion engine to improve accuracy, and adds new data to continuously improve the model's performance.

[1107] The above are the specific processing steps for carrying out the invention.

[1108] Example 2

[1109] 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."

[1110] Traditional marketing and sales strategies struggle to collect and analyze large amounts of data in real time to generate effective strategies. It's also difficult to provide personalized responses that take into account the user's emotional state. Furthermore, there's a lack of a mechanism for continuously improving models based on feedback, limiting the accuracy of the system.

[1111] 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: means for selecting information sources from users and accessing external APIs and internal storage devices; means for preprocessing the collected information and filtering unnecessary information; means for inputting the preprocessed information into a generative AI model and analyzing brand awareness, reputation, and target audience attributes; means for displaying the analysis results on a display device in real time and providing them to the user; means for predicting future trends and providing the prediction results; means for generating sales promotion and sales tactics based on the analysis and prediction results and notifying the user; means for obtaining user feedback and retraining the model; means for recognizing user emotions in real time using an emotion engine; and means for generating personalized strategies based on the user's emotional state and providing them to the user. This enables rapid and accurate understanding of market trends and individualized responses that take the user's emotional state into consideration. Furthermore, by continuously collecting feedback and improving the model, the accuracy of the system can be improved.

[1112] "User" refers to the entity that uses this system to execute marketing and sales strategies.

[1113] "Source" refers to an external API or internal storage device from which data is obtained.

[1114] "External API" refers to an application program interface for obtaining data from external services provided on the Internet.

[1115] "Internal storage device" refers to a database that stores past purchase history, customer information, etc., managed within a company.

[1116] "Collected Information" refers to data obtained by the Server from sources designated by the User.

[1117] "Preprocessing" refers to the process of filtering unnecessary information from collected information and standardizing the format of text data.

[1118] A "generative AI model" refers to an artificial intelligence model that can analyze data and make predictions in a generative way.

[1119] "Brand awareness" refers to an indicator of how familiar consumers are with a particular brand.

[1120] "Reputation" refers to the general evaluation or opinion of a brand or product.

[1121] A "target audience" refers to a group of customers who may be interested in a particular product or service.

[1122] "Display Device" refers to a device (e.g., dashboard, monitor) for visually displaying collected and analyzed data.

[1123] "Future trends" refers to predicted changes and trends in the market or brand in the future.

[1124] "Sales promotion" refers to the strategies and measures taken to promote a particular product or service in the market.

[1125] "Sales tactics" refer to the specific approaches and methods used to effectively sell a company's products or services.

[1126] "Feedback" refers to information provided to the server regarding the results of campaigns and strategies implemented by the user.

[1127] "Model retraining" refers to the process of improving a generative AI model based on feedback to improve its performance.

[1128] "Emotion engine" refers to a system component for identifying and analyzing user emotions in real time.

[1129] An "individualized strategy" refers to a strategy that provides optimal marketing and sales tactics to specific users based on their emotional state or specific attributes.

[1130] The system uses generative AI to understand brand awareness, reputation, and target audience insights in real time, combined with an emotion engine that recognizes user emotions to enhance marketing and sales strategies. The entire system is comprised of multiple servers, terminals, and users.

[1131] Hardware and Software Configuration

[1132] The server is primarily responsible for data collection, preprocessing, analysis, real-time feedback, prediction, strategy generation, and emotion recognition. The specific hardware used includes a network interface for processing API requests, a database server for storing data, and a high-performance GPU for running the generative AI model. The software includes the Twitter API, an SQL database, natural language processing libraries (e.g., NLTK and Spacy), and machine learning frameworks (e.g., TensorFlow and PyTorch).

[1133] The terminal is a computer or smartphone that users use to access the system and check analysis results and strategic proposals. The terminal connects to the server via a web browser or dedicated application.

[1134] Examples of data collection

[1135] The server collects data related to a specific brand by accessing external APIs (such as the Twitter API) or an internal database. For example, the server collects tweets containing "new drink name" and other related hashtags. At the same time, the server retrieves past purchase history and customer information from the internal database using SQL queries.

[1136] Example prompt sentence:

[1137] Get the latest tweets related to "New Drink Names".

[1138] Specific examples of data preprocessing

[1139] The server preprocesses the collected data, filtering out unnecessary information, removing spam messages and irrelevant data, and standardizing the format of text data. Specifically, it uses natural language processing technologies (e.g., NLTK and Spacy) to clean up and standardize the text format of tweets.

[1140] Example prompt sentence:

[1141] Please remove spam messages from your data and standardize the format.

[1142] Specific examples of data analysis

[1143] The server inputs the preprocessed data into a generative AI model (e.g., BERT or GPT-3) to analyze brand awareness, reputation, and target audience attributes. It also uses a machine learning model to perform sentiment analysis and classify the data as positive, negative, or neutral. It also generates a target audience profile based on user purchasing behavior and hobbies.

[1144] Example prompt sentence:

[1145] Analyze brand sentiment and categorize sentiment based on pre-processed data.

[1146] Examples of real-time feedback

[1147] The server displays the analysis results in real time on a dashboard and provides them to users. This uses WebSocket technology to continuously update the latest analysis results on the dashboard UI component, allowing users to grasp market trends in real time.

[1148] Example prompt sentence:

[1149] View the latest analytics on your dashboard.

[1150] Specific examples of future trend predictions

[1151] The server uses generative AI models to predict future trends for the brand. For example, the server creates a predictive model based on time-series data to forecast fluctuations in brand awareness and reputation over the next few months. The results are displayed on a dashboard.

[1152] Example prompt sentence:

[1153] Predict your brand awareness for the next three months.

[1154] Example of strategy generation

[1155] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user. Specifically, it makes strategic suggestions such as "We recommend offering a promotion code in the next campaign." These suggestions are not only displayed on the dashboard, but also sent to the user via email and push notifications.

[1156] Example prompt sentence:

[1157] Generate recommended strategies for your next campaign.

[1158] Specific examples of feedback

[1159] Users provide feedback on the results of their campaigns and strategies to the server. The server receives this feedback and retrains the AI ​​model to improve the accuracy of the system. For example, users input campaign results through a dashboard or dedicated app, and the server collects the data and uses it for the next model training.

[1160] Example prompt sentence:

[1161] Save the campaign results as feedback.

[1162] Examples of emotion engines

[1163] The server uses an emotion engine to recognize users' emotions in real time. It performs emotion analysis based on the user's online activities (comments, purchase history, access logs, etc.) to identify their current emotional state. For example, it calculates an emotion score based on the user's comments and purchase history and classifies them as positive, negative, or neutral.

[1164] Example prompt sentence:

[1165] Perform sentiment analysis on user comments.

[1166] Examples of personalized strategies

[1167] The server generates and provides personalized marketing and sales strategies based on the user's emotional state, promoting specific promotions in the case of a positive emotional state and providing customer support information in the case of a negative emotional state.

[1168] Example prompt sentence:

[1169] Generate strategies based on the user's emotional state.

[1170] Using this system, companies can quickly and accurately grasp market trends, implement effective marketing and sales strategies, and even provide personalized responses that take into account the user's emotional state.

[1171] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1172] Step 1: Data collection

[1173] The server collects data from information sources. Specifically, the server accesses an external API (e.g., Twitter API) and retrieves tweets containing specified keywords (e.g., "new beverage name"). The server also retrieves past purchase history and customer information from its internal storage device using SQL queries. The input here is the information source and keywords and conditions to be collected, and the output is the collected raw data.

[1174] Specific behavior:

[1175] The server periodically issues requests to the Twitter API and parses the API responses to retrieve data.

[1176] The server extracts the required information from the internal storage using SQL queries.

[1177] Step 2: Data Preprocessing

[1178] The server preprocesses the collected data and filters out unnecessary information. Specifically, the server uses natural language processing techniques (e.g., NLTK or Spacy) to clean up the tweet text data and standardize the format. The input here is the collected raw data, and the output is the preprocessed clean data.

[1179] Specific behavior:

[1180] The server cleans the text data, removing spam messages and irrelevant data.

[1181] The server uses regular expressions to remove certain patterns and converts the text data into a unified format.

[1182] Step 3: Data analysis

[1183] The server then inputs the preprocessed data into a generative AI model (e.g., BERT or GPT-3) to analyze brand awareness, reputation, and target audience attributes. The input is the preprocessed clean data, and the output is the analysis results. The server also uses a machine learning model to perform sentiment analysis and classify the data as positive, negative, or neutral.

[1184] Specific behavior:

[1185] The server inputs the preprocessed text into a generative AI model to obtain emotion scores and attribute information.

[1186] The server uses a sentiment analysis model to classify tweets and assign them to positive, negative, or neutral sentiment categories.

[1187] Step 4: Real-time feedback

[1188] The server displays the analysis results in real time on a display device and provides them to the user. The input here is the results of the data analysis, and the output is dashboard data that is updated in real time.

[1189] Specific behavior:

[1190] The server pushes data to the dashboard over a WebSocket connection.

[1191] The dashboard UI components receive the latest data and present it visually to the user.

[1192] Step 5: Predict future trends

[1193] The server uses a generative AI model to predict future trends for the brand, where the input is past data and current analysis results, and the output is predicted future trend data.

[1194] Specific behavior:

[1195] The server predicts the data using a time series prediction model (e.g., ARIMA or LSTM).

[1196] The server sends the prediction results to a dashboard for display to the user.

[1197] Step 6: Strategy Generation

[1198] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user, where the inputs are the analysis and prediction results and the output is the generated strategy proposals.

[1199] Specific behavior:

[1200] The server generates a strategy template based on the analysis and prediction results.

[1201] The server sends the generated strategy to the user via email or push notification.

[1202] Step 7: Gather feedback

[1203] The results of the campaigns and strategies executed by the user are provided as feedback to the server. The input here is the results of the campaigns and strategies, and the output is the collected feedback data. The server uses this feedback to retrain the AI ​​model.

[1204] Specific behavior:

[1205] Users enter campaign results through a dashboard or dedicated app.

[1206] The server collects the feedback data and uses it for the next model training.

[1207] Step 8: Emotion Recognition

[1208] The server uses an emotion engine to recognize the user's emotions in real time, where the input is the user's online activity data and the output is the recognized emotional state.

[1209] Specific behavior:

[1210] The server analyzes the user's comments and purchase history and calculates an emotion score.

[1211] The server appropriately tags and stores the emotional state.

[1212] Step 9: Personalized strategy generation

[1213] The server generates personalized marketing and sales strategies based on the user's emotional state and provides them to the user, where the input is the user's emotional state and the output is personalized strategy proposals.

[1214] Specific behavior:

[1215] The server classifies the user's emotional state and selects an appropriate strategy template.

[1216] The server generates a personalized strategy and notifies the user via email or push notification.

[1217] (Application example 2)

[1218] 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."

[1219] Product quality control is a critical issue in modern factory operations. Mass-produced products, in particular, face a high risk of quality variation, making quality control time-consuming and costly. Traditional quality control systems struggle to identify problems in real time and propose rapid solutions. Systems that can predict future quality trends and propose appropriate improvement measures are also lacking. This creates a demand for advanced systems that can maintain high levels of product quality while improving factory efficiency.

[1220] 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.

[1221] In this invention, the server includes: means for collecting product quality data from sensors and camera devices installed in the factory; means for preprocessing the collected product quality data and filtering incomplete or abnormal data; means for inputting the preprocessed quality data into a generative AI model to monitor, evaluate, and identify problems in real time; means for displaying analysis results in real time on a dashboard for factory workers and immediately notifying them if a problem occurs; means for predicting future quality trends based on past quality data and providing the prediction results to workers; and means for proposing optimal product inspection methods and manufacturing processes based on the analysis and prediction results. This enables highly accurate real-time monitoring and rapid response, not only enabling efficient product quality control but also enabling proactive improvement measures to be implemented based on future trend predictions.

[1222] A "data source" is a source of information from which information is obtained, such as an external API or an internal database.

[1223] "External API" refers to an interface to external services or data provided over the Internet.

[1224] An "internal database" is a system that stores data managed within an organization.

[1225] "Preprocessing" refers to the process of converting raw data into a form that is easier to analyze, such as removing noise and standardizing the format.

[1226] A "generative AI model" is a model that uses generative adversarial networks (GANs) or other AI algorithms to analyze data and generate new information.

[1227] A "brand" is a collective term for the elements that create a particular image and reputation for a company or product.

[1228] "Awareness" is a measure of how many people know about a brand or product.

[1229] "Reputation" refers to the evaluation and feelings that consumers and the market have about a brand or product.

[1230] "Target audience" refers to a specific group of consumers that you are targeting with marketing and advertising.

[1231] "Attributes" refer to characteristics of the target audience, such as gender, age, and hobbies.

[1232] "Real-time" refers to the state in which data is acquired and processed almost immediately.

[1233] A "dashboard" is an interface that visually displays data analysis results and important information.

[1234] "Future trends" refers to future tendencies and developments predicted by analyzing past and present data.

[1235] "Feedback" refers to information used to adjust future actions based on the results and reactions of implemented measures and strategies.

[1236] "Factory" refers to a place where products are produced or processed.

[1237] A "sensor" is a device that senses physical conditions or data and converts them into electrical signals.

[1238] A "camera device" is a device that captures video and converts it into digital data.

[1239] "Quality Data" refers to data relating to the characteristics and performance of a product.

[1240] "Incomplete data" refers to data that is missing or missing.

[1241] "Abnormal data" refers to data that exhibits unusual values ​​or patterns that fall outside of normal ranges.

[1242] "Monitoring" means constantly checking a specific object or condition.

[1243] "Evaluation" means judging the quality or performance of something.

[1244] "Notification" refers to the immediate sharing of specific information or results.

[1245] "Quality Trends" refers to trends or fluctuations in product quality.

[1246] "Improvement measures" refer to methods and means for solving current problems and improving the situation.

[1247] "Product inspection methods" refer to specific methods and techniques for checking product quality.

[1248] "Manufacturing process" refers to the series of steps taken to create a product from raw materials.

[1249] This invention provides a system for streamlining factory product quality control and continuously providing high-quality products. This system utilizes various sensors and camera devices installed in the factory to collect and analyze product quality data in real time. It also proposes quick countermeasures based on the analysis results and predicts future quality trends, thereby improving product quality.

[1250] System Configuration

[1251] The server operates in the following specific steps:

[1252] Data collection methods

[1253] The server collects product quality data using sensors and camera devices installed in the factory, including various quality indicators such as product dimensions, weight, and appearance. Specifically, it uses IoT devices and communication modules to acquire data from the various sensors and cameras via a network.

[1254] Data preprocessing measures

[1255] The server preprocesses the collected data, filtering out incomplete or abnormal data, and utilizes open-source natural language processing libraries (e.g., NLTK) and image processing libraries (e.g., OpenCV) to remove noise from the data and convert it into a unified format.

[1256] Data Analysis Methods

[1257] The preprocessed data is input into a generative AI model. The server uses the generative AI model to monitor product quality in real time and identify problems. A sentiment engine (e.g., Sentiment Analyzer) is also used for analysis, analyzing quality data and feedback logs from workers to identify the cause of quality issues.

[1258] Real-time feedback tools

[1259] The server displays the analysis results in real time on a dashboard for factory workers and immediately notifies them if any problems occur. WebSocket technology is used to continuously update the dashboard's UI components with the latest information, allowing workers to immediately detect problems and take action.

[1260] A means of predicting future trends

[1261] The server predicts future quality trends based on past quality data. To do this, it uses a time series analysis algorithm to predict future fluctuations in product quality. The analysis results are displayed on a dashboard and provided to workers.

[1262] Strategy Generation Method

[1263] Based on the analysis and prediction results, the server proposes optimal product inspection methods and manufacturing processes, allowing factory workers to quickly implement appropriate improvement measures.

[1264] Specific examples

[1265] For example, this system is used in a factory to control the quality of steel products. The server collects data on product dimensions, weight, and surface condition from sensors and camera devices, and evaluates the quality in real time. The preprocessed data is input into a generative AI model to detect quality anomalies. The detection results are displayed on a dashboard and workers are immediately notified. Furthermore, future quality trends are predicted based on past data, and if a prediction is made that the quality of steel products will improve with a certain probability next month, specific improvement measures are suggested to workers.

[1266] Prompt Sentence Examples

[1267] Input product quality data and previous day's manufacturing process data into a generative AI model and suggest your next quality control strategy. For example, are there any trends like the following?

[1268] By using the system of the present invention, not only can factory quality control be carried out efficiently, but proactive improvement measures can also be implemented based on predictions of future trends.

[1269] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1270] Step 1: Data collection

[1271] The server collects product quality data from sensors and camera devices installed in the factory. The sensors acquire physical data such as product dimensions, weight, and temperature, while the camera devices acquire images of the product's appearance. The input is sensor and camera data, and the output is the collected results of this data. Specifically, the server accesses the API of each device, requests data, and acquires it.

[1272] Step 2: Data Preprocessing

[1273] The server preprocesses the collected data and filters out incomplete or abnormal data. Specifically, it performs missing value imputation, outlier detection and removal, and data format standardization. The input is the data collected in step 1, and the output is quality data that has been filtered and cleaned. Specifically, it performs these processes using a natural language processing library (NLTK) and an image processing library (OpenCV).

[1274] Step 3: Data analysis

[1275] The server inputs the preprocessed data into the generative AI model to monitor, evaluate, and identify problems in real time with product quality. The input is preprocessed quality data, and the output is the quality evaluation results. Specifically, the server inputs data into the generative AI model and uses a sentiment engine (Sentiment Analyzer) to evaluate the quality.

[1276] Step 4: Real-time feedback

[1277] The server displays the analysis results in real time on a dashboard for factory workers and immediately notifies them if any problems occur. The dashboard is updated using WebSocket technology. The input is the quality evaluation results, and the output is real-time notifications to workers and the dashboard display. Specifically, the UI components are updated via the WebSocket connection.

[1278] Step 5: Predict future trends

[1279] The server predicts future quality trends based on past quality data. It uses a time series analysis algorithm to predict future quality fluctuations. The input is past and current quality data, and the output is the future quality trend prediction result. Specifically, it performs predictive analysis using a time series analysis algorithm.

[1280] Step 6: Strategy Generation

[1281] The server proposes optimal product inspection methods and manufacturing processes based on the analysis and prediction results. The inputs are prediction results and quality assessment results, and the output is improvement proposals. Specifically, the server notifies the dashboard and workers of the optimal strategy derived by the generative AI model.

[1282] Step 7: Get feedback and retrain

[1283] The server receives feedback on the implemented improvements and retrains the generative AI model. The input is feedback data from the workers, and the output is the retrained generative AI model. Specifically, the server uses the received feedback data to update the AI ​​model's parameters and improve its accuracy.

[1284] 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.

[1285] 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.

[1286] 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.

[1287] [Fourth embodiment]

[1288] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1289] 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.

[1290] 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).

[1291] 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.

[1292] 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.

[1293] 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).

[1294] 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.

[1295] 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.

[1296] 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.

[1297] 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.

[1298] 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.

[1299] 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.

[1300] 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."

[1301] This invention is a system that utilizes generative AI to understand brand awareness, reputation, and target audience insights in real time to enhance marketing and sales strategies. This system is composed of multiple servers, terminals, and users.

[1302] System Configuration

[1303] 1. Data Collection Methods

[1304] The server accesses external APIs and internal databases to collect data related to a specific brand. For example, the server uses the Twitter API to collect tweets containing the brand name and related hashtags. At the same time, it retrieves past purchase history and customer information from the internal database.

[1305] 2. Data preprocessing methods

[1306] The server preprocesses the collected data and filters out unnecessary information. Specifically, the server removes spam messages and irrelevant data and standardizes the format of the text data. During this process, the server analyzes the text data using natural language processing technology.

[1307] 3. Data Analysis Methods

[1308] The server inputs the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target audience attributes. At this stage, the server uses a machine learning model to perform sentiment analysis and classify the data as positive, negative, or neutral. It also generates a target audience profile based on user purchasing behavior and hobbies.

[1309] 4. Real-time feedback methods

[1310] The server displays the analysis results in real time on a dashboard and provides them to the user. For example, the server uses WebSocket technology to continuously update the latest analysis results to the dashboard UI component, allowing users to grasp market trends in real time.

[1311] 5. Means of predicting future trends

[1312] Sarver uses generative AI models to predict future trends for brands. Based on time-series data, it predicts fluctuations in awareness and reputation over the next few months and displays the results on a dashboard. For example, Sarver predicts that brand awareness will increase by 5% next month.

[1313] 6. Strategy Generation Method

[1314] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user. Specifically, the server makes strategy suggestions such as "recommending providing a promotion code in the next campaign." These suggestions are notified to the user not only via the dashboard but also by email and push notifications.

[1315] 7. Feedback channels

[1316] Users provide feedback on the results of their campaigns and strategies to the server. The server uses this feedback to retrain the AI ​​model and improve the accuracy of the system. For example, a user can input the results of a campaign, and the server can collect that data and use it for the next model training.

[1317] Specific examples

[1318] For example, a beverage manufacturer might use this system when launching a new product. The server collects tweets related to the "new beverage name" from social media and analyzes its popularity in real time. The server also combines this with past purchasing data to identify that "men in their 30s who love sports" are the primary target audience. The generative AI model then predicts that the product's popularity will increase next month. Based on this information, the system suggests a strategy to the user, such as "running a promotion during a sporting event," and provides feedback on the results to be used next time.

[1319] The above is a specific embodiment of the present invention. By using this system, companies can quickly and accurately grasp market trends and implement effective marketing and sales strategies.

[1320] The processing flow will be explained below.

[1321] Step 1: Select a data source

[1322] A data source is selected so that the server can collect data from an external API (e.g., Twitter API) or an internal database. The server obtains an API key and authentication information to establish access to the external API.

[1323] Step 2: Getting the data

[1324] The server sends a request containing specific keywords or hashtags to retrieve data from selected data sources. Specifically, the server sends an API request to search tweets and posts for the brand name and related keywords.

[1325] Step 3: Preprocessing the data

[1326] The server preprocesses the retrieved data and filters out unnecessary information, specifically filtering out spam messages and normalizing the text (e.g., converting to lowercase and removing extra whitespace). In this step, natural language processing techniques are used to analyze the text data.

[1327] Step 4: Save your data

[1328] The server stores the pre-processed data in an internal database for quick access in subsequent processing steps.

[1329] Step 5: Perform sentiment analysis

[1330] The server uses a generative AI model to perform sentiment analysis on the pre-processed data, for example, taking tweet or review text as input and classifying it as positive, negative, or neutral.

[1331] Step 6: Target Audience Analysis

[1332] The server combines the results of the sentiment analysis with the user's purchasing history to perform a detailed analysis of the target audience's attributes, specifically creating a profile of their age group, gender, hobbies, and preferences.

[1333] Step 7: Provide real-time feedback

[1334] The server displays the analysis results in a dashboard in real time, using WebSocket or API to ensure that data is reflected to users in real time.

[1335] Step 8: Perform future trend predictions

[1336] The server uses a generative AI model to predict future trends for the brand, predicting fluctuations in name recognition and reputation based on time-series data.

[1337] Step 9: Providing prediction results

[1338] The server displays the prediction results on a dashboard and provides them to the user, visualizing them in graphs and charts for easy understanding.

[1339] Step 10: Generate a marketing and sales strategy

[1340] Based on the analysis and prediction results, the server proposes specific marketing and sales strategies to the user, such as providing promotion codes and recommending targeted campaigns.

[1341] Step 11: Communicate your strategy

[1342] The server notifies the user of the generated marketing and sales strategies via the dashboard, email, and push notifications.

[1343] Step 12: Get feedback

[1344] Users report the results of their campaigns and strategies to the server, specifically by inputting the success rate and KPIs (key performance indicators) of the campaigns.

[1345] Step 13: Retrain the model

[1346] The server uses feedback data obtained from users to retrain the generative AI model to improve its accuracy, and adds new data to continuously improve the model's performance.

[1347] The above are the specific processing steps for carrying out the invention.

[1348] Example 1

[1349] 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."

[1350] Traditional methods for formulating marketing and sales strategies make it difficult to grasp real-time market trends, making it difficult to quickly derive effective strategies. Furthermore, there is a lack of a process for improving the system based on feedback, making it difficult to improve accuracy. Furthermore, existing systems require a great deal of time and effort to preprocess and analyze massive amounts of data, making efficient operation difficult.

[1351] 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.

[1352] In this invention, the server includes means for selecting a data source from a user and accessing an external API or an internal database, means for preprocessing the collected data and filtering out unnecessary information, means for inputting the preprocessed data into a generative AI model and analyzing brand awareness, reputation, and target audience attributes, means for displaying the analysis results on a dashboard in real time and providing them to the user, means for predicting future trends and providing the prediction results, means for generating a marketing and sales strategy using the generative AI model and notifying the user, and means for obtaining user feedback and retraining the model. This makes it possible to grasp market trends in real time and formulate highly accurate marketing and sales strategies.

[1353] A "data source" is a reference from which data is collected, such as an external API or an internal database.

[1354] "Preprocessing" is the process of filtering unnecessary information from collected data and standardizing the data format.

[1355] A "generative AI model" is a model that uses machine learning algorithms to analyze data and generate insights.

[1356] "Brand awareness" is an indicator of how well a particular brand is known among the general public.

[1357] "Reputation" refers to consumer opinions and ratings of a particular brand or product.

[1358] A "target audience" is a group of consumers that is the target of marketing and advertising.

[1359] "Displaying on the dashboard in real time" means that the analysis results are immediately reflected in the user interface so that the user can check them immediately.

[1360] "Predicting future trends" means predicting future market trends and fluctuations in brand awareness based on collected data.

[1361] A "marketing and sales strategy" is a strategic plan for increasing sales of a product or service.

[1362] "User notification" is the process of informing users of generated strategies and important information.

[1363] "Feedback" is any opinion or evaluation provided based on the results of an implemented strategy or campaign.

[1364] "Model retraining" is the re-learning process of using collected feedback data to improve the accuracy of an existing generative AI model.

[1365] This invention is a system that utilizes generative AI to understand brand awareness, reputation, and target audience insights in real time to enhance marketing and sales strategies. This system is composed of multiple servers, terminals, and users.

[1366] System Configuration

[1367] 1. Data Collection Methods

[1368] The server accesses external APIs and internal databases to collect data related to a specific brand. For example, the server uses the Twitter API to collect tweets containing a specific brand name or related hashtags. The server also retrieves past purchase history and customer information from internal databases. This data is obtained from databases such as MongoDB and MySQL.

[1369] 2. Data preprocessing methods

[1370] The server preprocesses the collected data and filters out unnecessary information. Specifically, the server removes spam messages and irrelevant data and standardizes the format of the text data. During this process, the server uses natural language processing techniques to analyze the text data using the Python libraries NLTK and spaCy.

[1371] 3. Data Analysis Methods

[1372] The server inputs the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target audience attributes. At this stage, the server performs sentiment analysis using the Hugging Face Transformer model to classify the data as positive, negative, or neutral. The server also uses a clustering algorithm (e.g., K-means) to generate a target audience profile based on user purchasing behavior and hobbies.

[1373] 4. Real-time feedback methods

[1374] The server displays the analysis results in real time on a dashboard and provides them to the user. For example, the server uses WebSocket technology to continuously update the dashboard's UI components with the latest analysis results. The front-end terminal uses React.js and D3.js to update graphs and charts based on the received data.

[1375] 5. Means of predicting future trends

[1376] The server uses a generative AI model to predict future trends for the brand. Based on time-series data, it uses a Long Short-Term Memory (LSTM) network to predict fluctuations in awareness and reputation over the next few months, and displays the results on a dashboard.

[1377] 6. Strategy Generation Method

[1378] The server uses the generative AI model to generate marketing and sales strategies and notify users. Specifically, the server uses the generative AI model to propose strategies such as "recommend offering a promotional code in the next campaign" based on demand forecasts. These strategies are not only displayed on the dashboard, but also notified to users via email and push notifications using AWS SNS (Simple Notification Service).

[1379] 7. Feedback channels

[1380] Users provide feedback to the server on the results of their campaigns and strategies. They input the results data using their devices and send it to the server. The server then retrains the generative AI model based on the collected feedback data, improving the accuracy of the next analysis and prediction. TensorFlow and PyTorch are used for retraining.

[1381] Specific examples

[1382] For example, a beverage manufacturer might use this system when launching a new product. The server collects tweets related to the "new beverage name" from social media and analyzes its popularity in real time. The server also combines this with past purchasing data to identify that "men in their 30s who love sports" are the primary target audience. The generative AI model then predicts that the product's popularity will increase next month. Based on this information, the system suggests a strategy to the user, such as "running a promotion during a sporting event," and provides feedback on the results to be used next time.

[1383] (Example of a prompt)

[1384] Collect data to analyze your brand awareness and reputation.

[1385] Preprocess the collected data to filter out irrelevant data.

[1386] Analyze the preprocessed data and classify it as positive, negative, or neutral.

[1387] Predict your brand's future reputation and visibility.

[1388] Based on the analysis, suggest a strategy for your next campaign.

[1389] The above is a specific embodiment of the present invention. By using this system, companies can quickly and accurately grasp market trends and implement effective marketing and sales strategies.

[1390] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1391] Step 1: Data source selection and data collection

[1392] 1. Input: The user selects a data source containing brand names and related keywords.

[1393] 2. Processing: The server accesses external APIs and internal databases based on the input information received from the user. Specifically, it uses the Twitter API to collect tweets containing the specified brand name or hashtag. It also retrieves past purchase history and customer information from internal databases (e.g., MongoDB or MySQL).

[1394] 3. Output: The collected data is stored in the server and is ready for the next step.

[1395] Step 2: Data Preprocessing

[1396] 1. Input: Data collected in step 1.

[1397] 2. Processing: The server preprocesses the data, removing spam messages and irrelevant data and standardizing the format of the text data. Specifically, it uses the Python libraries NLTK and spaCy to parse the text data and convert the format.

[1398] 3. Output: Preprocessed and clean data is generated and sent to the next step.

[1399] Step 3: Input and analysis into the generative AI model

[1400] 1. Input: The preprocessed data from step 2.

[1401] 2. Processing: The server inputs this data into a generative AI model. It uses Hugging Face's Transformer model to perform sentiment analysis and classify the data as positive, negative, or neutral. It also uses a clustering algorithm (e.g., K-means) to generate a profile of the target audience.

[1402] 3. Output: The analyzed data is generated and results are obtained regarding awareness, reputation, and target audience attributes.

[1403] Step 4: Real-time feedback

[1404] 1. Input: The analysis results generated in step 3.

[1405] 2. Processing: The server reflects the results in real time on the dashboard. Using WebSocket technology, the latest analysis results are updated sequentially on the dashboard's UI components. The terminal visualizes the data using React.js and D3.js.

[1406] 3. Output: The latest data is displayed in the dashboard that users access.

[1407] Step 5: Predict future trends

[1408] 1. Input: Analysis results and time series data generated in Step 3.

[1409] 2. Processing: The server uses a Long Short-Term Memory (LSTM) network to predict future fluctuations in popularity and reputation.

[1410] 3. Output: The prediction results are generated and displayed on a dashboard.

[1411] Step 6: Strategy Generation

[1412] 1. Input: The analytical and prediction results generated in Step 3 and Step 5.

[1413] 2. Processing: The server uses the generative AI model to generate marketing and sales strategies, such as providing promotional codes for the next campaign.

[1414] 3. Output: The generated strategies are displayed on the dashboard and notified to the user via email and push notifications using AWS SNS (Simple Notification Service).

[1415] Step 7: Get feedback and retrain the model

[1416] 1. Input: The user inputs the results of the campaigns and strategies they have executed using a terminal and sends them to the server.

[1417] 2. Processing: The server uses the collected feedback data to retrain the generative AI model using TensorFlow or PyTorch.

[1418] 3. Output: A retrained AI model is generated, improving the accuracy of the next analysis or prediction.

[1419] (Application example 1)

[1420] 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."

[1421] It has been difficult to accurately understand the effectiveness of traditional advertising campaigns in real time and adjust strategies immediately. It has also been difficult to generate detailed profiles of target users based on their purchasing behavior and hobbies. Furthermore, there has been a lack of efficient means to process large amounts of data obtained from social media and provide sentiment analysis results in real time.

[1422] 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.

[1423] In this invention, the server includes means for selecting information sources from users and accessing external APIs and internal databases, means for preprocessing the collected data and filtering unnecessary information, means for inputting the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target user attributes, means for displaying sentiment analysis results collected from social media on a dashboard in real time and providing them to users, means for predicting future trends and providing the prediction results, means for generating marketing and sales strategies based on the analysis results and prediction results and notifying users, and means for obtaining user feedback and retraining the model. This makes it possible to accurately grasp the effectiveness of advertising campaigns in real time and immediately adjust strategies.

[1424] "User" means any person or entity that uses this system.

[1425] A "source" is a service or system that provides data, such as an external API or an internal database.

[1426] A "generative AI model" is an artificial intelligence model that is trained to take data as input and analyze brand awareness, reputation, and target user attributes.

[1427] "Social media" refers to online platforms such as Twitter where users post and share information.

[1428] "Sentiment analysis" is the process of classifying the sentiment of text data into positive, negative, or neutral.

[1429] A "dashboard" is an interface for displaying analysis results and prediction results in real time.

[1430] A "marketing strategy" is a plan or method aimed at promoting a product or service.

[1431] A "sales strategy" refers to the plans and methods of sales activities aimed at increasing sales.

[1432] "Feedback" refers to information received from users regarding the results of implemented campaigns and strategies.

[1433] "Retraining" is the process of retraining an artificial intelligence model based on feedback data.

[1434] "Future trends" refers to future trends predicted based on time series data and generative AI models.

[1435] This invention is a system that utilizes generative AI models to understand brand awareness, reputation, and target user insights in real time to enhance marketing and sales strategies. The system consists of a server and an application that can be accessed from a smartphone or other device. A specific example is shown below.

[1436] Functional Configuration

[1437] 1. Data Collection Methods

[1438] The server selects the information source specified by the user and accesses external APIs (e.g., Twitter API) or internal databases. The server collects data containing the specified brand name and related hashtags. This data includes social media posts, the user's past purchase history, customer information, etc.

[1439] 2. Data preprocessing methods

[1440] The server preprocesses the collected data and filters out unnecessary information, using natural language processing techniques to remove spam messages and irrelevant data and standardize the format of the text data.

[1441] 3. Data Analysis Methods

[1442] The preprocessed data is input into a generative AI model, and the server analyzes brand awareness, reputation, and target user attributes. A machine learning model is used to perform sentiment analysis and classify the data as positive, negative, or neutral. It also generates target user profiles based on user purchasing behavior and hobbies.

[1443] 4. Real-time feedback methods

[1444] The server displays the analysis results in real time on a dashboard and provides them to users. WebSocket technology is used to continuously update the dashboard UI components with the latest analysis results, allowing users to grasp market trends in real time.

[1445] 5. Means of predicting future trends

[1446] The server uses generative AI models to predict future trends, using time-series data to forecast fluctuations in popularity and reputation over the next few months, and displays the results on a dashboard.

[1447] 6. Strategy Generation Method

[1448] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user of them, specifically proposing the content and timing of promotions to be carried out in the next campaign.

[1449] 7. Feedback channels

[1450] Users provide feedback on the results of their campaigns and strategies to the server, which then uses this feedback to retrain the AI ​​model and improve the system's accuracy, enabling it to provide more effective strategies for the next campaign.

[1451] Examples of specific examples and prompts

[1452] For example, a beverage manufacturer might use this system when running an advertising campaign for a new product. The server collects tweets related to the "new beverage name" from social media and analyzes their popularity in real time. The server also combines this with past purchasing data to identify that "men in their 30s who love sports" are the primary target users. Based on this information, the generative AI model predicts that "product popularity will increase next month" and suggests a strategy to users: "run a promotion during a sporting event." When the results of the campaign are fed back, the AI ​​model is retrained using that data.

[1453] Example prompt sentence:

[1454] """

[1455] Your goal is to write a program that tracks social media sentiment related to a particular brand and analyzes that sentiment. You will use the Twitter API for data collection and natural language processing techniques for data preprocessing and sentiment analysis. You will set up a web server using Flask to display the results in real time.

[1456] """

[1457] Using this system, companies can quickly and accurately grasp market trends and implement effective marketing and sales strategies.

[1458] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1459] Step 1:

[1460] The server selects information sources specified by the user and collects data. Specifically, when the user specifies a specific brand name or related hashtag, the server collects social media posts via the Twitter API. It also accesses an internal database to obtain related purchase history and customer information. The input is the specified brand name and hashtag, and the output is the collected social media posts and customer data.

[1461] Step 2:

[1462] The server preprocesses the collected data and filters out unnecessary information. Specifically, it uses natural language processing techniques to remove spam messages and irrelevant data and standardize the format of the text data. This process involves, for example, using automated scripts to remove URLs and mentions and remove stop words. The input is the collected social media posts and customer data, and the output is the preprocessed data.

[1463] Step 3:

[1464] The preprocessed data is input into a generative AI model to analyze brand awareness, reputation, and target user attributes. The server uses a machine learning model to perform sentiment analysis on the text data and classify the data as positive, negative, or neutral. It also generates a target user profile based on the user's purchasing behavior and hobbies. The input is the preprocessed data, and the output is the sentiment analysis results and the target user profile.

[1465] Step 4:

[1466] The server displays the analysis results in real time on a dashboard and provides them to users. WebSocket technology is used to continuously update the latest analysis results to the dashboard's UI components, allowing users to grasp market trends in real time. The input is the sentiment analysis results and target user profiles, and the output is real-time analysis information displayed on the dashboard.

[1467] Step 5:

[1468] The server uses a generative AI model to predict future trends. Based on time-series data, it predicts fluctuations in popularity and reputation over the next few months and displays the results on a dashboard. Specifically, it uses a predictive algorithm to calculate future data points and visualizes the results. The input is time-series data, and the output is a forecast of future trends.

[1469] Step 6:

[1470] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user. Specifically, it proposes the content and timing of promotions to be carried out in the next campaign. The server sends the generated strategy information to the user via email or push notification. The input is the analysis and prediction results, and the output is a proposed marketing and sales strategy.

[1471] Step 7:

[1472] Users provide feedback to the server on the results of their implemented campaigns and strategies. The server uses this feedback to retrain the AI ​​model and improve the accuracy of the system. For example, a user inputs the effectiveness of a campaign, and the server collects that data and uses it for the next model training. The input is the user's feedback data, and the output is a retrained AI model.

[1473] 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.

[1474] This invention is a system that utilizes generative AI to grasp brand awareness, reputation, and target audience insights in real time, and combines it with an emotion engine that recognizes user emotions to enhance marketing and sales strategies. This system is composed of multiple servers, terminals, and users.

[1475] System Configuration

[1476] 1. Data Collection Methods

[1477] The server accesses external APIs and internal databases to collect data related to a specific brand. For example, the server uses the Twitter API to collect tweets containing the brand name and related hashtags. At the same time, it retrieves past purchase history and customer information from the internal database.

[1478] 2. Data preprocessing methods

[1479] The server preprocesses the collected data and filters out unnecessary information. Specifically, the server removes spam messages and irrelevant data and standardizes the format of the text data. During this process, the server analyzes the text data using natural language processing technology.

[1480] 3. Data Analysis Methods

[1481] The server inputs the preprocessed data into a generative AI model to analyze brand awareness, reputation, and target audience attributes. The server then uses a machine learning model to perform sentiment analysis and classify the data as positive, negative, or neutral. It also generates a target audience profile based on user purchasing behavior and hobbies.

[1482] 4. Real-time feedback methods

[1483] The server displays the analysis results in real time on a dashboard and provides them to the user. For example, the server uses WebSocket technology to continuously update the latest analysis results to the dashboard UI component, allowing users to grasp market trends in real time.

[1484] 5. Means of predicting future trends

[1485] Sarver uses generative AI models to predict future trends for brands. Based on time-series data, it predicts fluctuations in awareness and reputation over the next few months and displays the results on a dashboard. For example, Sarver predicts that brand awareness will increase by 5% next month.

[1486] 6. Strategy Generation Method

[1487] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user. Specifically, the server makes strategy suggestions such as "recommending providing a promotion code in the next campaign." These suggestions are notified to the user not only via the dashboard but also by email and push notifications.

[1488] 7. Feedback channels

[1489] Users provide feedback on the results of their campaigns and strategies to the server. The server uses this feedback to retrain the AI ​​model and improve the accuracy of the system. For example, a user can input the results of a campaign, and the server can collect that data and use it for the next model training.

[1490] 8. Emotional Engine Means

[1491] The server uses an emotion engine to recognize the user's emotions in real time. For example, the server performs emotion analysis based on data acquired from the user's online activities (comments, purchase history, access logs, etc.) to identify the user's current emotional state.

[1492] 9. Personalized Strategy Generation

[1493] The server generates and provides personalized marketing and sales strategies to the user based on the user's emotional state, for example, promoting a specific promotion if the user is in a positive emotional state and providing customer support information if the user is in a negative emotional state.

[1494] Specific examples

[1495] For example, a beverage manufacturer could use this system when launching a new product. The server collects related tweets on the "new beverage name" from social media and analyzes their popularity in real time. The server also combines this with past purchasing data to identify that "sports-loving men in their 30s" are the primary target audience. The generative AI model then predicts that the product's popularity will increase next month. Furthermore, the emotion engine analyzes the emotional state of users from their comments and feedback and provides personalized strategies. For example, if there are a lot of positive reactions, a limited promotional campaign can be promoted, and if there are a lot of negative reactions, immediate customer support can be provided.

[1496] The above is a specific embodiment of the present invention. By using this system, companies can quickly and accurately grasp market trends, implement effective marketing and sales strategies, and even provide personalized responses that take into account the emotional state of users.

[1497] The processing flow will be explained below.

[1498] Step 1: Select a data source

[1499] The server selects a data source to allow it to collect data from an external API (e.g., Twitter API) or an internal database. The server obtains an API key and authentication information to establish access to the external API.

[1500] Step 2: Getting the data

[1501] The server sends a request containing specific keywords or hashtags to retrieve data from selected data sources. Specifically, the server sends an API request to search tweets and posts for the brand name and related keywords.

[1502] Step 3: Preprocessing the data

[1503] The server preprocesses the retrieved data and filters out unnecessary information, specifically filtering out spam messages and normalizing the text (e.g., converting to lowercase and removing extra whitespace). In this step, natural language processing techniques are used to analyze the text data.

[1504] Step 4: Save your data

[1505] The server stores the pre-processed data in an internal database for quick access in subsequent processing steps.

[1506] Step 5: Perform sentiment analysis

[1507] The server uses a generative AI model to perform sentiment analysis on the pre-processed data, for example, taking tweet or review text as input and classifying it as positive, negative, or neutral.

[1508] Step 6: Target Audience Analysis

[1509] The server combines the results of the sentiment analysis with the user's purchasing history to perform a detailed analysis of the target audience's attributes, specifically creating a profile of their age group, gender, hobbies, and preferences.

[1510] Step 7: Provide real-time feedback

[1511] The server displays the analysis results in a dashboard in real time, using WebSocket or API to ensure that data is reflected to users in real time.

[1512] Step 8: Perform future trend predictions

[1513] The server uses a generative AI model to predict future trends for the brand, predicting fluctuations in name recognition and reputation based on time-series data.

[1514] Step 9: Providing prediction results

[1515] The server displays the prediction results on a dashboard and provides them to the user, visualizing them in graphs and charts for easy understanding.

[1516] Step 10: Generate a marketing and sales strategy

[1517] Based on the analysis and prediction results, the server proposes specific marketing and sales strategies to the user, such as providing promotion codes and recommending targeted campaigns.

[1518] Step 11: Communicate your strategy

[1519] The server notifies the user of the generated marketing and sales strategies via the dashboard, email, and push notifications.

[1520] Step 12: Collect user emotion data

[1521] The server collects data on the user's emotional state from their online activities (comments, purchase history, access logs, etc.) and uses an emotion engine to analyze the collected data in real time to identify the user's emotional state.

[1522] Step 13: Individualized strategy generation

[1523] The server generates and provides personalized marketing and sales strategies based on the user's emotional state, for example, promoting new product introductions and promotional campaigns to users in a positive emotional state, and providing customer support information to users in a negative emotional state.

[1524] Step 14: Getting feedback

[1525] Users report the results of their campaigns and strategies to the server, specifically by inputting the success rate and KPIs (key performance indicators) of the campaigns.

[1526] Step 15: Retrain the model

[1527] The server uses feedback data obtained from users to retrain the generative AI model and emotion engine to improve accuracy, and adds new data to continuously improve the model's performance.

[1528] The above are the specific processing steps for carrying out the invention.

[1529] Example 2

[1530] 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."

[1531] Traditional marketing and sales strategies struggle to collect and analyze large amounts of data in real time to generate effective strategies. It's also difficult to provide personalized responses that take into account the user's emotional state. Furthermore, there's a lack of a mechanism for continuously improving models based on feedback, limiting the accuracy of the system.

[1532] 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: means for selecting information sources from users and accessing external APIs and internal storage devices; means for preprocessing the collected information and filtering unnecessary information; means for inputting the preprocessed information into a generative AI model and analyzing brand awareness, reputation, and target audience attributes; means for displaying the analysis results on a display device in real time and providing them to the user; means for predicting future trends and providing the prediction results; means for generating sales promotion and sales tactics based on the analysis and prediction results and notifying the user; means for obtaining user feedback and retraining the model; means for recognizing user emotions in real time using an emotion engine; and means for generating personalized strategies based on the user's emotional state and providing them to the user. This enables rapid and accurate understanding of market trends and individualized responses that take the user's emotional state into consideration. Furthermore, by continuously collecting feedback and improving the model, the accuracy of the system can be improved.

[1533] "User" refers to the entity that uses this system to execute marketing and sales strategies.

[1534] "Source" refers to an external API or internal storage device from which data is obtained.

[1535] "External API" refers to an application program interface for obtaining data from external services provided on the Internet.

[1536] "Internal storage device" refers to a database that stores past purchase history, customer information, etc., managed within a company.

[1537] "Collected Information" refers to data obtained by the Server from sources designated by the User.

[1538] "Preprocessing" refers to the process of filtering unnecessary information from collected information and standardizing the format of text data.

[1539] A "generative AI model" refers to an artificial intelligence model that can analyze data and make predictions in a generative way.

[1540] "Brand awareness" refers to an indicator of how familiar consumers are with a particular brand.

[1541] "Reputation" refers to the general evaluation or opinion of a brand or product.

[1542] A "target audience" refers to a group of customers who may be interested in a particular product or service.

[1543] "Display Device" refers to a device (e.g., dashboard, monitor) for visually displaying collected and analyzed data.

[1544] "Future trends" refers to predicted changes and trends in the market or brand in the future.

[1545] "Sales promotion" refers to the strategies and measures taken to promote a particular product or service in the market.

[1546] "Sales tactics" refer to the specific approaches and methods used to effectively sell a company's products or services.

[1547] "Feedback" refers to information provided to the server regarding the results of campaigns and strategies implemented by the user.

[1548] "Model retraining" refers to the process of improving a generative AI model based on feedback to improve its performance.

[1549] "Emotion engine" refers to a system component for identifying and analyzing user emotions in real time.

[1550] An "individualized strategy" refers to a strategy that provides optimal marketing and sales tactics to specific users based on their emotional state or specific attributes.

[1551] The system uses generative AI to understand brand awareness, reputation, and target audience insights in real time, combined with an emotion engine that recognizes user emotions to enhance marketing and sales strategies. The entire system is comprised of multiple servers, terminals, and users.

[1552] Hardware and Software Configuration

[1553] The server is primarily responsible for data collection, preprocessing, analysis, real-time feedback, prediction, strategy generation, and emotion recognition. The specific hardware used includes a network interface for processing API requests, a database server for storing data, and a high-performance GPU for running the generative AI model. The software includes the Twitter API, an SQL database, natural language processing libraries (e.g., NLTK and Spacy), and machine learning frameworks (e.g., TensorFlow and PyTorch).

[1554] The terminal is a computer or smartphone that users use to access the system and check analysis results and strategic proposals. The terminal connects to the server via a web browser or dedicated application.

[1555] Examples of data collection

[1556] The server collects data related to a specific brand by accessing external APIs (such as the Twitter API) or an internal database. For example, the server collects tweets containing "new drink name" and other related hashtags. At the same time, the server retrieves past purchase history and customer information from the internal database using SQL queries.

[1557] Example prompt sentence:

[1558] Get the latest tweets related to "New Drink Names".

[1559] Specific examples of data preprocessing

[1560] The server preprocesses the collected data, filtering out unnecessary information, removing spam messages and irrelevant data, and standardizing the format of text data. Specifically, it uses natural language processing technologies (e.g., NLTK and Spacy) to clean up and standardize the text format of tweets.

[1561] Example prompt sentence:

[1562] Please remove spam messages from your data and standardize the format.

[1563] Specific examples of data analysis

[1564] The server inputs the preprocessed data into a generative AI model (e.g., BERT or GPT-3) to analyze brand awareness, reputation, and target audience attributes. It also uses a machine learning model to perform sentiment analysis and classify the data as positive, negative, or neutral. It also generates a target audience profile based on user purchasing behavior and hobbies.

[1565] Example prompt sentence:

[1566] Analyze brand sentiment and categorize sentiment based on pre-processed data.

[1567] Examples of real-time feedback

[1568] The server displays the analysis results in real time on a dashboard and provides them to users. This uses WebSocket technology to continuously update the latest analysis results on the dashboard UI component, allowing users to grasp market trends in real time.

[1569] Example prompt sentence:

[1570] View the latest analytics on your dashboard.

[1571] Specific examples of future trend predictions

[1572] The server uses generative AI models to predict future trends for the brand. For example, the server creates a predictive model based on time-series data to forecast fluctuations in brand awareness and reputation over the next few months. The results are displayed on a dashboard.

[1573] Example prompt sentence:

[1574] Predict your brand awareness for the next three months.

[1575] Example of strategy generation

[1576] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user. Specifically, it makes strategic suggestions such as "We recommend offering a promotion code in the next campaign." These suggestions are not only displayed on the dashboard, but also sent to the user via email and push notifications.

[1577] Example prompt sentence:

[1578] Generate recommended strategies for your next campaign.

[1579] Specific examples of feedback

[1580] Users provide feedback on the results of their campaigns and strategies to the server. The server receives this feedback and retrains the AI ​​model to improve the accuracy of the system. For example, users input campaign results through a dashboard or dedicated app, and the server collects the data and uses it for the next model training.

[1581] Example prompt sentence:

[1582] Save the campaign results as feedback.

[1583] Examples of emotion engines

[1584] The server uses an emotion engine to recognize users' emotions in real time. It performs emotion analysis based on the user's online activities (comments, purchase history, access logs, etc.) to identify their current emotional state. For example, it calculates an emotion score based on the user's comments and purchase history and classifies them as positive, negative, or neutral.

[1585] Example prompt sentence:

[1586] Perform sentiment analysis on user comments.

[1587] Examples of personalized strategies

[1588] The server generates and provides personalized marketing and sales strategies based on the user's emotional state, promoting specific promotions in the case of a positive emotional state and providing customer support information in the case of a negative emotional state.

[1589] Example prompt sentence:

[1590] Generate strategies based on the user's emotional state.

[1591] Using this system, companies can quickly and accurately grasp market trends, implement effective marketing and sales strategies, and even provide personalized responses that take into account the user's emotional state.

[1592] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1593] Step 1: Data collection

[1594] The server collects data from information sources. Specifically, the server accesses an external API (e.g., Twitter API) and retrieves tweets containing specified keywords (e.g., "new beverage name"). The server also retrieves past purchase history and customer information from its internal storage device using SQL queries. The input here is the information source and keywords and conditions to be collected, and the output is the collected raw data.

[1595] Specific behavior:

[1596] The server periodically issues requests to the Twitter API and parses the API responses to retrieve data.

[1597] The server extracts the required information from the internal storage using SQL queries.

[1598] Step 2: Data Preprocessing

[1599] The server preprocesses the collected data and filters out unnecessary information. Specifically, the server uses natural language processing techniques (e.g., NLTK or Spacy) to clean up the tweet text data and standardize the format. The input here is the collected raw data, and the output is the preprocessed clean data.

[1600] Specific behavior:

[1601] The server cleans the text data, removing spam messages and irrelevant data.

[1602] The server uses regular expressions to remove certain patterns and converts the text data into a unified format.

[1603] Step 3: Data analysis

[1604] The server then inputs the preprocessed data into a generative AI model (e.g., BERT or GPT-3) to analyze brand awareness, reputation, and target audience attributes. The input is the preprocessed clean data, and the output is the analysis results. The server also uses a machine learning model to perform sentiment analysis and classify the data as positive, negative, or neutral.

[1605] Specific behavior:

[1606] The server inputs the preprocessed text into a generative AI model to obtain emotion scores and attribute information.

[1607] The server uses a sentiment analysis model to classify tweets and assign them to positive, negative, or neutral sentiment categories.

[1608] Step 4: Real-time feedback

[1609] The server displays the analysis results in real time on a display device and provides them to the user. The input here is the results of the data analysis, and the output is dashboard data that is updated in real time.

[1610] Specific behavior:

[1611] The server pushes data to the dashboard over a WebSocket connection.

[1612] The dashboard UI components receive the latest data and present it visually to the user.

[1613] Step 5: Predict future trends

[1614] The server uses a generative AI model to predict future trends for the brand, where the input is past data and current analysis results, and the output is predicted future trend data.

[1615] Specific behavior:

[1616] The server predicts the data using a time series prediction model (e.g., ARIMA or LSTM).

[1617] The server sends the prediction results to a dashboard for display to the user.

[1618] Step 6: Strategy Generation

[1619] The server generates marketing and sales strategies based on the analysis and prediction results and notifies the user, where the inputs are the analysis and prediction results and the output is the generated strategy proposals.

[1620] Specific behavior:

[1621] The server generates a strategy template based on the analysis and prediction results.

[1622] The server sends the generated strategy to the user via email or push notification.

[1623] Step 7: Gather feedback

[1624] The results of the campaigns and strategies executed by the user are provided as feedback to the server. The input here is the results of the campaigns and strategies, and the output is the collected feedback data. The server uses this feedback to retrain the AI ​​model.

[1625] Specific behavior:

[1626] Users enter campaign results through a dashboard or dedicated app.

[1627] The server collects the feedback data and uses it for the next model training.

[1628] Step 8: Emotion Recognition

[1629] The server uses an emotion engine to recognize the user's emotions in real time, where the input is the user's online activity data and the output is the recognized emotional state.

[1630] Specific behavior:

[1631] The server analyzes the user's comments and purchase history and calculates an emotion score.

[1632] The server appropriately tags and stores the emotional state.

[1633] Step 9: Personalized strategy generation

[1634] The server generates personalized marketing and sales strategies based on the user's emotional state and provides them to the user, where the input is the user's emotional state and the output is personalized strategy proposals.

[1635] Specific behavior:

[1636] The server classifies the user's emotional state and selects an appropriate strategy template.

[1637] The server generates a personalized strategy and notifies the user via email or push notification.

[1638] (Application example 2)

[1639] 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."

[1640] Product quality control is a critical issue in modern factory operations. Mass-produced products, in particular, face a high risk of quality variation, making quality control time-consuming and costly. Traditional quality control systems struggle to identify problems in real time and propose rapid solutions. Systems that can predict future quality trends and propose appropriate improvement measures are also lacking. This creates a demand for advanced systems that can maintain high levels of product quality while improving factory efficiency.

[1641] 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.

[1642] In this invention, the server includes: means for collecting product quality data from sensors and camera devices installed in the factory; means for preprocessing the collected product quality data and filtering incomplete or abnormal data; means for inputting the preprocessed quality data into a generative AI model to monitor, evaluate, and identify problems in real time; means for displaying analysis results in real time on a dashboard for factory workers and immediately notifying them if a problem occurs; means for predicting future quality trends based on past quality data and providing the prediction results to workers; and means for proposing optimal product inspection methods and manufacturing processes based on the analysis and prediction results. This enables highly accurate real-time monitoring and rapid response, not only enabling efficient product quality control but also enabling proactive improvement measures to be implemented based on future trend predictions.

[1643] A "data source" is a source of information from which information is obtained, such as an external API or an internal database.

[1644] "External API" refers to an interface to external services or data provided over the Internet.

[1645] An "internal database" is a system that stores data managed within an organization.

[1646] "Preprocessing" refers to the process of converting raw data into a form that is easier to analyze, such as removing noise and standardizing the format.

[1647] A "generative AI model" is a model that uses generative adversarial networks (GANs) or other AI algorithms to analyze data and generate new information.

[1648] A "brand" is a collective term for the elements that create a particular image and reputation for a company or product.

[1649] "Awareness" is a measure of how many people know about a brand or product.

[1650] "Reputation" refers to the evaluation and feelings that consumers and the market have about a brand or product.

[1651] "Target audience" refers to a specific group of consumers that you are targeting with marketing and advertising.

[1652] "Attributes" refer to characteristics of the target audience, such as gender, age, and hobbies.

[1653] "Real-time" refers to the state in which data is acquired and processed almost immediately.

[1654] A "dashboard" is an interface that visually displays data analysis results and important information.

[1655] "Future trends" refers to future tendencies and developments predicted by analyzing past and present data.

[1656] "Feedback" refers to information used to adjust future actions based on the results and reactions of implemented measures and strategies.

[1657] "Factory" refers to a place where products are produced or processed.

[1658] A "sensor" is a device that senses physical conditions or data and converts them into electrical signals.

[1659] A "camera device" is a device that captures video and converts it into digital data.

[1660] "Quality Data" refers to data relating to the characteristics and performance of a product.

[1661] "Incomplete data" refers to data that is missing or missing.

[1662] "Abnormal data" refers to data that exhibits unusual values ​​or patterns that fall outside of normal ranges.

[1663] "Monitoring" means constantly checking a specific object or condition.

[1664] "Evaluation" means judging the quality or performance of something.

[1665] "Notification" refers to the immediate sharing of specific information or results.

[1666] "Quality Trends" refers to trends or fluctuations in product quality.

[1667] "Improvement measures" refer to methods and means for solving current problems and improving the situation.

[1668] "Product inspection methods" refer to specific methods and techniques for checking product quality.

[1669] "Manufacturing process" refers to the series of steps taken to create a product from raw materials.

[1670] This invention provides a system for streamlining factory product quality control and continuously providing high-quality products. This system utilizes various sensors and camera devices installed in the factory to collect and analyze product quality data in real time. It also proposes quick countermeasures based on the analysis results and predicts future quality trends, thereby improving product quality.

[1671] System Configuration

[1672] The server operates in the following specific steps:

[1673] Data collection methods

[1674] The server collects product quality data using sensors and camera devices installed in the factory, including various quality indicators such as product dimensions, weight, and appearance. Specifically, it uses IoT devices and communication modules to acquire data from the various sensors and cameras via a network.

[1675] Data preprocessing measures

[1676] The server preprocesses the collected data, filtering out incomplete or abnormal data, and utilizes open-source natural language processing libraries (e.g., NLTK) and image processing libraries (e.g., OpenCV) to remove noise from the data and convert it into a unified format.

[1677] Data Analysis Methods

[1678] The preprocessed data is input into a generative AI model. The server uses the generative AI model to monitor product quality in real time and identify problems. A sentiment engine (e.g., Sentiment Analyzer) is also used for analysis, analyzing quality data and feedback logs from workers to identify the cause of quality issues.

[1679] Real-time feedback tools

[1680] The server displays the analysis results in real time on a dashboard for factory workers and immediately notifies them if any problems occur. WebSocket technology is used to continuously update the dashboard's UI components with the latest information, allowing workers to immediately detect problems and take action.

[1681] A means of predicting future trends

[1682] The server predicts future quality trends based on past quality data. To do this, it uses a time series analysis algorithm to predict future fluctuations in product quality. The analysis results are displayed on a dashboard and provided to workers.

[1683] Strategy Generation Method

[1684] Based on the analysis and prediction results, the server proposes optimal product inspection methods and manufacturing processes, allowing factory workers to quickly implement appropriate improvement measures.

[1685] Specific examples

[1686] For example, this system is used in a factory to control the quality of steel products. The server collects data on product dimensions, weight, and surface condition from sensors and camera devices, and evaluates the quality in real time. The preprocessed data is input into a generative AI model to detect quality anomalies. The detection results are displayed on a dashboard and workers are immediately notified. Furthermore, future quality trends are predicted based on past data, and if a prediction is made that the quality of steel products will improve with a certain probability next month, specific improvement measures are suggested to workers.

[1687] Prompt Sentence Examples

[1688] Input product quality data and previous day's manufacturing process data into a generative AI model and suggest your next quality control strategy. For example, are there any trends like the following?

[1689] By using the system of the present invention, not only can factory quality control be carried out efficiently, but proactive improvement measures can also be implemented based on predictions of future trends.

[1690] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1691] Step 1: Data collection

[1692] The server collects product quality data from sensors and camera devices installed in the factory. The sensors acquire physical data such as product dimensions, weight, and temperature, while the camera devices acquire images of the product's appearance. The input is sensor and camera data, and the output is the collected results of this data. Specifically, the server accesses the API of each device, requests data, and acquires it.

[1693] Step 2: Data Preprocessing

[1694] The server preprocesses the collected data and filters out incomplete or abnormal data. Specifically, it performs missing value imputation, outlier detection and removal, and data format standardization. The input is the data collected in step 1, and the output is quality data that has been filtered and cleaned. Specifically, it performs these processes using a natural language processing library (NLTK) and an image processing library (OpenCV).

[1695] Step 3: Data analysis

[1696] The server inputs the preprocessed data into the generative AI model to monitor, evaluate, and identify problems in real time with product quality. The input is preprocessed quality data, and the output is the quality evaluation results. Specifically, the server inputs data into the generative AI model and uses a sentiment engine (Sentiment Analyzer) to evaluate the quality.

[1697] Step 4: Real-time feedback

[1698] The server displays the analysis results in real time on a dashboard for factory workers and immediately notifies them if any problems occur. The dashboard is updated using WebSocket technology. The input is the quality evaluation results, and the output is real-time notifications to workers and the dashboard display. Specifically, the UI components are updated via the WebSocket connection.

[1699] Step 5: Predict future trends

[1700] The server predicts future quality trends based on past quality data. It uses a time series analysis algorithm to predict future quality fluctuations. The input is past and current quality data, and the output is the future quality trend prediction result. Specifically, it performs predictive analysis using a time series analysis algorithm.

[1701] Step 6: Strategy Generation

[1702] The server proposes optimal product inspection methods and manufacturing processes based on the analysis and prediction results. The inputs are prediction results and quality assessment results, and the output is improvement proposals. Specifically, the server notifies the dashboard and workers of the optimal strategy derived by the generative AI model.

[1703] Step 7: Get feedback and retrain

[1704] The server receives feedback on the implemented improvements and retrains the generative AI model. The input is feedback data from the workers, and the output is the retrained generative AI model. Specifically, the server uses the received feedback data to update the AI ​​model's parameters and improve its accuracy.

[1705] 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.

[1706] 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.

[1707] 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 robot 414.

[1708] 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.

[1709] 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.

[1710] 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.

[1711] 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).

[1712] 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.

[1713] 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 emoti...

Claims

1. A means for users to select data sources and access external APIs and internal databases, A means of preprocessing the collected data and filtering out unnecessary information; A means to input pre-processed data into a generative AI model to analyze brand awareness, reputation, and target audience attributes; A means to display the analysis results in real time on a dashboard and provide them to users, A means of predicting future trends and providing forecast results; A means for generating and informing users of marketing and sales strategies based on the analytical and predictive results; A means to obtain user feedback and retrain the model; A system including:

2. 10. The system of claim 1, further comprising means for classifying the analysis results into positive, negative, or neutral sentiment.

3. 10. The system of claim 1, further comprising means for performing detailed analysis of the hobbies and purchasing behavior of the target audience.

Images