system

Abstract

Provide a system. 【Solution means】 An information collection means for acquiring information from a plurality of information sources via a communication network, A preprocessing means for filtering unnecessary information from the acquired information, A generation model learning means for learning a generation model based on past success examples using the filtered data, An information generation means for generating new and interesting content using the learned generation model, An information disclosure means for publicly releasing the generated content on an appropriate online platform, A communication means for interacting and communicating with users to promote engagement, An analysis means for analyzing the trends and engagement of followers after posting, A system including the above.

Description

Technical Field

[0004] , ,

[0005] , , ,

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, and includes steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance as a response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] As online marketing activities increase, a promotion method using influencers with a large number of followers has attracted attention. However, in the conventional method, a high cost is incurred when hiring an influencer, and it is difficult for companies lacking know - how to increase followers on their own to utilize it. Therefore, there is a need for a technology to efficiently acquire followers on SNS at low cost and conduct marketing activities.

Means for Solving the Problems

[0005] This invention provides a system that enables efficient follower acquisition and effective marketing activities by utilizing an AI agent that acts as an influencer on social media. Specifically, data is acquired from online platforms by an information gathering means, and the data is filtered and cleansed by a preprocessing means. A generative model learning means learns a generative model based on past successes, and an information generation means automatically generates attractive posts. An information publication means publishes these generated posts, and a communication means handles user interaction. Finally, an analysis means analyzes follower growth and engagement data to help inform future marketing strategies. In this way, even companies without expertise can maximize the effectiveness of their social media marketing at a low cost.

[0006] "Information gathering means" refers to devices that have the function of obtaining relevant data and posts from online platforms and information sources.

[0007] "Preprocessing means" refers to a system that performs filtering and cleansing on collected data to prepare it for analysis and model training.

[0008] A "generative model learning tool" is a device that uses filtered data to train a generative model based on past successes.

[0009] An "information generation means" is a device that has the function of generating new, attention-grabbing content using a pre-trained generative model.

[0010] "Information disclosure means" refers to a system that has the function of publishing generated content on an appropriate online platform.

[0011] "Communication means" refers to anything that has the function of facilitating interaction and communication with users and promoting their engagement.

[0012] "Analysis tools" are those that analyze the behavior and engagement of followers after a post and provide data to help inform future marketing strategies. [Brief explanation of the drawing]

[0013] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14]It is a sequence diagram showing the processing flow of a data processing system in Application Example 2 when a sentiment engine is combined.

Embodiment for Carrying Out the Invention

[0014] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0015] First, the terms used in the following description will be explained.

[0016] In the following embodiments, a processor with a reference number (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), etc.

[0017] In the following embodiments, a RAM (Random Access Memory) with a reference number is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0018] In the following embodiments, a storage with a reference number is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0019] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0020] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0021] [First Embodiment]

[0022] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0023] As shown in Figure 1, the 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.

[0024] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0025] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0026] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0027] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0028] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0029] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0031] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0032] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0033] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0034] This invention is a system for conducting effective marketing activities on social media using an AI agent. This system consists of three main elements: a server, a terminal, and a user. The processing of each element and specific examples are shown below.

[0035] Server Role

[0036] The server plays a central role in collecting, processing, generating, and analyzing information. First, the server collects data from designated online platforms. This includes data related to past successful posts and current trends. Next, the data acquired through information gathering is preprocessed to prepare it for training generative models. At this stage, it is crucial to remove spam data and noise to complete the cleaning process.

[0037] Subsequently, the generative model training system performs model training using pre-processed data. This generative model analyzes past successful cases and acquires patterns to maximize user engagement. After training is complete, the information generation system utilizes this model to automatically generate effective posts. The content thus generated is then reviewed and optimized by the server.

[0038] Terminal role

[0039] The terminal's primary role is to publish generated content on selected social media platforms based on instructions sent from the server. The published posts utilize content created by a generation AI, taking into account factors such as time of day and hashtags. This optimizes reach to the target user base.

[0040] User roles

[0041] Users interact with the information and content provided by this system. Engagement can be increased by having agents automatically respond to user comments via communication channels. Data such as user responses and changes in follower counts are sent to the server via analytical tools and used for future marketing activities.

[0042] Specific example

[0043] For example, when promoting a new product, the server analyzes the latest market trends and generates highly relevant content. This content, such as "the new product's distinctive features" and "suggested usage scenarios," is posted to social media via the user's device. Users view these posts, comment, and react, which are immediately analyzed by the server and reflected in the generation of subsequent content. This process allows companies to quickly and cost-effectively increase product awareness and promote sales.

[0044] In the above configuration, the present invention functions as a system that can efficiently acquire followers on social media and implement effective marketing activities.

[0045] The following describes the processing flow.

[0046] Step 1:

[0047] The server uses the SNS platform's API to collect relevant post data based on specific keywords and hashtags. A filter is applied to list posts that have previously received high engagement.

[0048] Step 2:

[0049] The server preprocesses the collected data. This process removes duplicate and spam data, cleans the text, filters out irrelevant noise, and generates a dataset with the necessary information organized.

[0050] Step 3:

[0051] The server uses pre-processed data to train a generative model. Here, natural language processing techniques are used to extract features from successful posts and learn patterns that lead to high engagement. These learning results are then incorporated into the generative model.

[0052] Step 4:

[0053] The server uses a pre-trained model to automatically generate new posts aligned with specific themes. This includes introductions to new products and information about specific campaigns. The generated content is then verified to ensure that its formatting and presentation are optimized.

[0054] Step 5:

[0055] The device receives instructions from the server and publishes the generated post to the appropriate social networking platform. The post is published at the optimal time based on the algorithm and user trends of the selected platform.

[0056] Step 6:

[0057] Users view posts on social media and react to them. Engagements such as comments and "likes" indicate the user's level of engagement with the post and influence subsequent server processing.

[0058] Step 7:

[0059] The server collects performance data for each post and analyzes engagement metrics (e.g., follower growth rate, likes, shares). This analysis is then used to improve future post creation and marketing strategies.

[0060] (Example 1)

[0061] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0062] The problem this invention aims to solve is to automate everything from data collection to content generation, publication, and user engagement optimization in order to conduct effective marketing activities on social media. In particular, it is necessary to process large amounts of data in real time and use natural language processing to filter and normalize it, thereby enabling the generation of effective content based on lean information. Furthermore, it is necessary to review and optimize the generated content and provide automated responses to maximize user engagement.

[0063] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0064] In this invention, the server includes information gathering means, preprocessing means, generative model learning means, information generation means, information publication means, communication means, analysis means, means for reviewing and optimizing the generated content, and means for providing automated responses to maximize user engagement. This enables efficient and effective marketing activities on social media, real-time data processing, optimization of information through natural language processing, and interaction with users.

[0065] An "information gathering device" is a device that has the function of acquiring information from multiple sources via a communication network and collecting data in real time.

[0066] A "preprocessing device" is a device that filters out unnecessary information from acquired data and then normalizes the data using natural language processing.

[0067] A "generative model learning device" is a device that uses pre-processed data to train a generative AI model and acquire patterns to maximize user engagement.

[0068] An "information generation means" is a device that automatically generates new information using a pre-trained generative model and creates content for posting.

[0069] "Information disclosure means" refers to a device that publishes generated content on a designated online platform and posts it considering the optimal time and hashtags.

[0070] A "communication device" is a device used to send and receive data and instructions in real time between a server and a terminal.

[0071] An "analysis tool" is a device that analyzes data such as user responses and changes in follow-up activity to extract information that can be used to improve future marketing activities.

[0072] "Means for reviewing and optimizing generated content" refers to a system that builds an AI-driven feedback loop to automatically improve the quality of generated content.

[0073] "Means of providing automated responses to maximize user engagement" refers to a device that automatically provides appropriate responses to user comments, thereby deepening interaction with the user.

[0074] This invention is an advanced automated system for effectively conducting marketing activities on social media using an AI agent. This system consists of three main elements: a server, a terminal, and a user. The roles and specific usage methods of each are described below.

[0075] Server Role

[0076] The server plays a central role in everything from information gathering to processing, generation, and analysis. First, the server uses information gathering tools to acquire data from multiple sources via the communication network. This data includes past posts, current trends, and the activities of competitors. Next, the server uses Python's Pandas and natural language processing tools as preprocessing tools to filter and normalize the data and remove unnecessary information.

[0077] Subsequently, the system learns patterns in the collected data using generative model learning methods such as TENSORFLOW® or PyTorch. This learning process explores the optimal methods for increasing user engagement. This trained model is then used as an information generation tool, automatically generating content using a text generation model (e.g., GPT). Furthermore, the server reviews the generated content and improves its quality through an AI feedback loop for optimization.

[0078] Terminal role

[0079] The terminal is responsible for publishing generated content to the SNS platform based on instructions sent from the server. The terminal receives data in real time using the WebSocket protocol and posts content considering the optimal time and hashtags. This allows for efficient reach to the target user base.

[0080] User roles

[0081] Users view posts through their devices and can comment, like, and share them. This interaction is fed back to the server via communication methods and analyzed to inform future marketing activities.

[0082] Examples of specific cases and prompt statements

[0083] For example, in the promotion of a new product, a server analyzes the latest market trends and generates highly relevant content. This content is posted on social media in a format that suggests the product's unique features and usage scenarios. Users react to this content, and this data is analyzed and used to generate subsequent content.

[0084] An example of a prompt might be: "Generate a social media post about the next new product based on the latest market trends. Maximize follower engagement."

[0085] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0086] Step 1:

[0087] The server retrieves data from multiple sources using information gathering tools. The input consists of data provided by each source. This data includes social media posts, trend information, and competitor activities. APIs are used to collect information in real time during data acquisition. The output is a collection of the retrieved raw data.

[0088] Step 2:

[0089] The server cleans the raw data obtained using preprocessing methods. The input is the raw data obtained in step 1. Specifically, the Pandas library in Python is used to filter out spam data and noise and normalize the data. Natural language processing tools are also used to format the text data. The output is refined, clean data.

[0090] Step 3:

[0091] The server uses a generative model training method to train a model on clean data. The input is the purified data obtained in step 2. This training is performed using machine learning libraries such as TensorFlow and PyTorch, and explores patterns that enhance user engagement. The output of this process is a trained generative AI model.

[0092] Step 4:

[0093] The server utilizes information generation tools and generates new content using a trained generative model. The input consists of the trained model and prompt text obtained in step 3. Specifically, a text generation model (e.g., GPT) automatically generates content for posting based on the instructions in the prompt text. This output becomes the text data to be posted.

[0094] Step 5:

[0095] The device publishes the content generated via the information disclosure mechanism on the SNS platform. The input is the generated content obtained in step 4. The device sends the content using the WebSocket protocol, taking into account the appropriate time of day and relevant hashtags. The output is the post published on SNS.

[0096] Step 6:

[0097] Users view published content and leave comments and reactions. Input is posts on social media. User interactions are fed back to the server, which is used for future data collection and content generation. Output is user response data.

[0098] Step 7:

[0099] The server uses analytical tools to analyze user response data and gain insights to improve future marketing strategies. The input is the user response data obtained in step 6. The analysis evaluates the effectiveness of the content using metrics such as follower growth / decrease and engagement rate. The output is a proposal for an improved marketing strategy.

[0100] (Application Example 1)

[0101] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0102] In today's advertising industry, there is a demand for personalized ads that are displayed at the right time based on users' interests and preferences. However, traditional advertising systems struggle to generate and display ads that respond to user behavior in real time, which limits the effectiveness of marketing activities.

[0103] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0104] In this invention, the server includes an information gathering device, a data processing device, and a generative model learning device. This makes it possible to acquire user behavior data in real time and generate and display advertising content optimized for the user based on that data.

[0105] An "information gathering device" is a device used to acquire data from individual information sources via an information network.

[0106] A "data processing device" is a device that analyzes collected data and transforms it into an optimized form based on user behavior and interests.

[0107] A "generative model learning device" is a device that uses processed data to build and train a learning model for generating advertising content based on user interests.

[0108] An "information generation device" is a device that has the function of actually generating advertisement copy and content using a trained model.

[0109] An "information transmission device" is a device that delivers generated advertising content to users through a user interface.

[0110] A "mutual communication device" is a device that transmits user responses and feedback to a server and uses them as data for generating the next content.

[0111] A "data analysis device" is a device that analyzes user response data and provides information for optimizing marketing strategies.

[0112] A "user interface device" is a device that allows users to directly interact with advertising content visually.

[0113] A "personalized display device" is a device that displays individually optimized advertisements to users according to their needs and interests.

[0114] To realize this invention, the system has the following configuration: The server operates with an information gathering device, a data processing device, and a generative model learning device. The information gathering device is responsible for collecting user data from social networking services (SNS) and other information sources via an information network. The data processing device cleanses the collected data and organizes it appropriately based on the user's interests and behavior. This process includes filtering spam data and removing unnecessary information.

[0115] Next, the generative model learning system uses the cleansed data and machine learning libraries such as TensorFlow to train a model for generating advertising content optimized for user interests. This enables the generation of new patterns based on past successful advertisements.

[0116] The terminal has a user interface device and is responsible for displaying generated advertising content to the user. When a user views an advertisement using the terminal, their response is sent to the server via a communication device. This response is analyzed in real time by the server's data analysis device and used to generate content for the next time.

[0117] For example, if a user frequently views sports-related posts, the generative model will use that data to create advertisements for the latest sports equipment. A possible prompt might be, "Based on the following information, create an ad copy that is likely to interest the user: The user's profile shows a lot of sports activity and they have liked many running-related posts." In this way, more appropriate and effective marketing activities become possible.

[0118] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0119] Step 1:

[0120] The server uses information gathering devices to collect user-related data from the information network. This data includes information about the user's activity history and interests on social networking services (SNS). It uses data obtained through SNS APIs as input and outputs it in a standardized format.

[0121] Step 2:

[0122] The server uses a data processing unit to cleanse the collected data. This process filters out spam and irrelevant information and extracts characteristic data based on user behavior. The input is the collected raw data, and the output is clean, analyzable data.

[0123] Step 3:

[0124] The server uses a generative model learning system to train a generative AI model using cleansed data and tools such as TensorFlow. In this step, patterns are identified to optimize user engagement. The input is the analyzed data obtained in step 2, and the output is the trained AI model.

[0125] Step 4:

[0126] Using the generated model, the server generates personalized advertising content tailored to the user's interests through an information generator. Here, ad copy is created using prompts. The input is a trained AI model and prompts, and the output is the specific ad content.

[0127] Step 5:

[0128] The terminal uses a user interface device to display generated advertising content to the user. This display is done via a smartphone or smart glasses. The input is advertising content received from the server, and the output is the advertisement visually presented to the user.

[0129] Step 6:

[0130] When a user interacts with an advertisement via their device, their response is sent to the server via a communication device. This response includes the number of clicks on the advertisement and comments. The input is user behavior data, and the output is feedback data for future content generation.

[0131] Step 7:

[0132] The server uses data analysis equipment to analyze user feedback and use it to improve advertising strategies. These analysis results are then incorporated into the creation of new advertising content. The input is feedback data, and the output is information for improving advertising strategies.

[0133] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0134] This invention incorporates an emotion engine that recognizes user emotions into a system that utilizes an AI agent to conduct effective marketing activities on social media. This allows for real-time analysis of the user's emotional state and optimization of marketing activities based on this analysis. The specific functions and usage examples for each element—server, terminal, and user—are described below.

[0135] Server Role

[0136] The server acts as the central hub of the system, handling data collection, processing, generation, and analysis. It also utilizes an emotion engine to analyze users' emotional states. Initially, the server collects posted data from social media platforms using various information gathering methods. By focusing on collecting data from particularly popular posts, comments, and reactions, it obtains a wealth of data regarding users' emotions.

[0137] Next, the data is cleaned using a preprocessing mechanism to organize it into a state suitable for training the generative model. Unnecessary data is removed during this process. The generative model training mechanism then performs model training incorporating sentiment data based on the preprocessed data. This enables content generation that takes into account changes in the user's emotions.

[0138] The emotional data acquired by the emotion engine helps generate appropriate content for users through information generation methods. For example, it can create encouraging messages for users who show positive reactions and supportive posts for users who show negative reactions.

[0139] Terminal role

[0140] The device publishes content generated from the server to the social networking platform at the appropriate time. This includes scheduling and adjusting posts based on sentiment analysis results. By posting during times when users are most active, engagement can be maximized.

[0141] User roles

[0142] Users interact with content provided through social media platforms. User comments and reactions are analyzed in real time using an emotion engine, and feedback is immediately provided by agents. Understanding each user's individual emotional state enables personalized responses.

[0143] Specific example

[0144] For example, during a campaign announcing a product launch, the server learns typical user response patterns from relevant historical data and continuously monitors users' real-time reactions using an emotion engine. In this case, it automatically generates posts offering coupons to users with positive emotions and provides special support guidance to users with negative emotions, thereby improving the overall user experience.

[0145] Thus, the present invention is a system that can realize more personalized marketing activities that take user emotions into consideration.

[0146] The following describes the processing flow.

[0147] Step 1:

[0148] The server collects post data related to predefined keywords and hashtags through the SNS platform's API. It prioritizes collecting posts that have garnered particularly high attention and stores this data in a database.

[0149] Step 2:

[0150] The server preprocesses the collected data. This process filters out unnecessary information and spam, extracting important data. It also cleans the text necessary for sentiment analysis and reduces noise.

[0151] Step 3:

[0152] The server trains a generative model using pre-processed data, while simultaneously using a sentiment engine to analyze sentiment data for each post and comment. The sentiment engine distinguishes between positive, negative, and neutral sentiments, tags them, and stores them.

[0153] Step 4:

[0154] Based on generative model training, the server automatically generates new posts that reflect sentiment data. Here, messages are adjusted according to the user's emotions, and content is constructed with an appropriate tone.

[0155] Step 5:

[0156] The system optimizes the timing of when the device receives content generated from the server and publishes it to the social networking platform. During this process, the posting schedule is adjusted based on the user's activity time and past posting patterns.

[0157] Step 6:

[0158] Users comment on and react to posts on social media, and their emotional states are recorded within the system. User feedback is reflected in real time and analyzed by the system's emotion engine.

[0159] Step 7:

[0160] The server analyzes user reaction data to posts and generates engagement metrics. This data is used to adjust future content creation and marketing strategies, optimizing the entire system.

[0161] (Example 2)

[0162] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0163] In today's digital society, effective marketing activities that take user emotions into account require the rapid and accurate collection of data from diverse sources and the selection of useful information from that data. However, traditional methods are inefficient in data collection and analysis, making it difficult to grasp user emotions in real time and reflect them in marketing strategies.

[0164] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0165] In this invention, the server includes an information acquisition means, a numerical processing means, and an emotion analysis means. This makes it possible to quickly collect necessary information from a large amount of data on an SNS platform, effectively remove unnecessary data, and then accurately analyze the user's emotions. This enables personalized marketing activities that respond to changes in the user's emotions.

[0166] "Information acquisition means" refers to a mechanism for obtaining necessary data from multiple information sources using wide-area communication.

[0167] "Numerical processing means" refers to functions that remove noise and unnecessary data from acquired data and prepare it in a format suitable for analysis.

[0168] "Computational model learning methods" refer to the process of training machine learning models using pre-processed data to obtain insights that are useful for future predictions and generation.

[0169] "Information creation means" refers to methods for generating new information or content based on a learned model.

[0170] "Information distribution means" refers to the function of publishing generated information to the appropriate platform at the optimal time.

[0171] "Information transmission means" refers to communication circuits and protocols used to exchange data within and outside a system.

[0172] "Evaluation methods" refer to methods for analyzing the performance of generated information and the overall system to identify areas for improvement.

[0173] "Emotional analysis tools" are functions that extract the user's emotional state from data and classify and analyze it.

[0174] This invention is a system based on advanced information technology that analyzes user emotions on a social networking service (SNS) platform and enables personalized marketing based on that analysis. The system is composed of an information acquisition means, a numerical processing means, an emotion analysis means, a computational model learning means, an information creation means, an information distribution means, and an information transmission means.

[0175] The server first acquires a large amount of user data from SNS platforms and other sources via wide-area communication using information acquisition methods. Specifically, it collects data primarily from posts, comments, and reactions that attract particularly high attention through API access. Based on this data, the server performs preprocessing using numerical processing methods to remove noise and unnecessary data, preparing it for analysis. Open-source libraries (e.g., NLTK and spaCy) can be used as natural language processing tools.

[0176] Next, the server analyzes the sentiment of the pre-processed data using sentiment analysis tools. Existing natural language processing APIs (e.g., Google® Cloud Natural Language API) are used for sentiment analysis, classifying the user's emotional state with tags such as positive, negative, and neutral. Based on these analysis results, a generative AI model is trained using computational model learning tools. TensorFlow and PyTorch are expected to be used as machine learning frameworks.

[0177] The server uses the acquired emotion data to generate content that responds to the user's emotions using an information creation mechanism. For example, it is possible to input a prompt such as, "Generate an encouraging message based on the user's positive response data," and the AI ​​can dynamically generate a message.

[0178] The generated content is published to social media platforms via information distribution methods at the optimal time determined by the device. The device analyzes user activity patterns and adjusts posts to coincide with the time when engagement is maximized. This can enhance the overall effectiveness of marketing activities.

[0179] Users react to and comment on the delivered content. This feedback is sent back to the server via communication channels and used for sentiment analysis and content generation in the next round. The learning process of the generative AI model is continuously improved by incorporating user feedback, enabling the delivery of more effective marketing messages.

[0180] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0181] Step 1:

[0182] The server uses information acquisition methods to obtain large amounts of user data from SNS platforms via wide-area communication. The input is raw posted data via API, and the output is a collection of raw data organized in chronological order. At this stage, priority is given to collecting posts with a large number of "likes" and "shares."

[0183] Step 2:

[0184] The server performs preprocessing on the acquired raw data using numerical processing tools. The input is the raw data obtained in step 1, and the output is the cleaned-up data. This preprocessing includes removing noise data and spam posts, and tokenizing comments and posts. The text is morphologically analyzed using a natural language processing library and formatted into a format suitable for analysis.

[0185] Step 3:

[0186] The server uses sentiment analysis tools to analyze pre-processed data and classify the user's emotional state. The input is cleaned-up data, and the output is data tagged with sentiment (positive, negative, neutral). A dedicated API is used for sentiment analysis, and accuracy is improved by scoring the intensity of each emotion.

[0187] Step 4:

[0188] The server uses computational model learning methods to train a generative AI model based on emotion-tagged data. The input is emotion-tagged data, and the output is a model that can predict changes in the user's emotions. A machine learning framework is used to extract features from the data and create an optimized model.

[0189] Step 5:

[0190] The server generates content using a trained model and prompt text via an information creation mechanism. The input is the trained model and the prompt text "Generate an encouraging message based on the user's positive response data," and the output is a personalized message. The AI ​​suggests the most appropriate wording based on the user's past response patterns.

[0191] Step 6:

[0192] The device publishes generated content to social media platforms at the optimal time via an information distribution method. The input is the generated content, and the output is a successful posting message to social media. The device analyzes past user activity times and delivers content at the time when engagement is highest.

[0193] Step 7:

[0194] Users provide feedback by reacting to and commenting on published content. The input is the content published on social media, and the output is new comments and reactions from users. This feedback is sent back to the server and used for sentiment analysis and content generation in the future.

[0195] (Application Example 2)

[0196] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0197] In marketing activities on social media, there is a challenge in that personalized ad delivery that takes into account the individual emotions of users is not being adequately implemented. As a result, ads that do not match users' interests or emotions are often delivered, which can lead to decreased ad effectiveness and a decline in the quality of the user experience.

[0198] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0199] In this invention, the server includes information gathering means, sentiment analysis means, and advertisement generation means. This makes it possible to analyze the user's emotions based on data collected from the SNS platform and deliver advertisements that are appropriate to the user's emotions in real time based on the analysis results.

[0200] "Information gathering means" refers to a system that has the function of acquiring information from multiple sources via a communication network.

[0201] A "preprocessing means" is a device that has the function of filtering out unnecessary information from the acquired information and using the refined information.

[0202] A "generative model learning tool" is a device that has the function of enabling an AI model to learn based on pre-processed data.

[0203] An "information generation means" is a device that has the function of generating appropriate content based on user sentiment data.

[0204] "Information disclosure means" refers to a system that has the functionality to publish generated content on a social networking service (SNS) platform.

[0205] "Communication means" refers to anything that has the function of sending and receiving data between each element within a system.

[0206] "Analysis tools" are those that analyze collected data and have the function of revealing users' emotions and behavioral patterns.

[0207] An "emotion analysis tool" is a device that has the function of analyzing the user's emotions from the collected data.

[0208] An "advertising generation method" is a device that has the function of generating advertisements appropriate to the user based on the results of sentiment analysis.

[0209] The server uses information gathering means to collect posted data from SNS platforms. In this process, it obtains the latest user posts and reaction data from Twitter and Facebook APIs via a communication network. This collected data is then filtered and refined by preprocessing means. Based on this refined data, a generative model learning means performs user sentiment analysis using an AI model (e.g., BERT or GPT).

[0210] The results of the sentiment analysis are used by the ad generation system to generate ads that are appropriate to the user's emotions. These ads are then prepared for posting to social media platforms via the information generation system and published at the appropriate time by the information publication system.

[0211] The device publishes this advertisement on social media based on instructions from the server and displays it at the appropriate time to increase user engagement. Users react to the published advertisement, which collects even more sentiment data that is used as training data for the entire system.

[0212] For example, if a user posts "Today is the best day!" on social media, the server analyzes this as a positive emotion and generates a restaurant coupon for a new product using an ad generation tool. Ultimately, this ad will appear in the user's feed.

[0213] Example prompt: "For posts that show positive user emotions, display ads that enhance enjoyment; for posts that show negative emotions, display ads that soothe the soul."

[0214] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0215] Step 1:

[0216] The server uses information gathering tools to obtain posted data and reaction data from the SNS platform's API via the communication network. In this process, messages and comments generated by users on SNS are collected as input, and the data is output in a standard data format for use in subsequent data processing.

[0217] Step 2:

[0218] The server uses preprocessing to filter out unnecessary data from the data collected in step 1 and extract the necessary information. In this process, since the input data may contain noise or duplicate data, these are removed, and clean data in a format suitable for sentiment analysis is output.

[0219] Step 3:

[0220] The server passes the refined data to a generative model training tool, which then uses an AI model (e.g., BERT or GPT) to analyze the user's emotions. This data calculation outputs a user's emotion score (positive, negative, neutral) based on the input clean data.

[0221] Step 4:

[0222] The server receives the output from the sentiment analysis system and uses the ad generation system to create ads that match the user's emotions. Here, the user's sentiment score is used as input, and highly relevant ad campaigns and promotional information are output.

[0223] Step 5:

[0224] The terminal uses information generation means to prepare to post the advertisement generated in step 4 to the SNS platform. In this preparation stage, the input advertisement data is converted into an appropriate format and output with scheduling information added.

[0225] Step 6:

[0226] The device publishes advertisements generated at the appropriate time on social media through an information disclosure mechanism. Here, posts prepared using the social media API are used as input data, timely publication processing is performed, and the result of the advertisement appearing in the user's feed is output.

[0227] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0228] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0229] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0230] [Second Embodiment]

[0231] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0232] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0233] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0234] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0235] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0236] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0237] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0238] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0239] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0240] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0241] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0242] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0243] This invention is a system for conducting effective marketing activities on social media using an AI agent. This system consists of three main elements: a server, a terminal, and a user. The processing of each element and specific examples are shown below.

[0244] Server Role

[0245] The server plays a central role in collecting, processing, generating, and analyzing information. First, the server collects data from designated online platforms. This includes data related to past successful posts and current trends. Next, the data acquired through information gathering is preprocessed to prepare it for training generative models. At this stage, it is crucial to remove spam data and noise to complete the cleaning process.

[0246] Subsequently, the generative model training system performs model training using pre-processed data. This generative model analyzes past successful cases and acquires patterns to maximize user engagement. After training is complete, the information generation system utilizes this model to automatically generate effective posts. The content thus generated is then reviewed and optimized by the server.

[0247] Terminal role

[0248] The terminal's primary role is to publish generated content on selected social media platforms based on instructions sent from the server. The published posts utilize content created by a generation AI, taking into account factors such as time of day and hashtags. This optimizes reach to the target user base.

[0249] User roles

[0250] Users interact with the information and content provided by this system. Engagement can be increased by having agents automatically respond to user comments via communication channels. Data such as user responses and changes in follower counts are sent to the server via analytical tools and used for future marketing activities.

[0251] Specific example

[0252] For example, when promoting a new product, the server analyzes the latest market trends and generates highly relevant content. This content, such as "the new product's distinctive features" and "suggested usage scenarios," is posted to social media via the user's device. Users view these posts, comment, and react, which are immediately analyzed by the server and reflected in the generation of subsequent content. This process allows companies to quickly and cost-effectively increase product awareness and promote sales.

[0253] In the above configuration, the present invention functions as a system that can efficiently acquire followers on social media and implement effective marketing activities.

[0254] The following describes the processing flow.

[0255] Step 1:

[0256] The server uses the SNS platform's API to collect relevant post data based on specific keywords and hashtags. A filter is applied to list posts that have previously received high engagement.

[0257] Step 2:

[0258] The server preprocesses the collected data. This process removes duplicate and spam data, cleans the text, filters out irrelevant noise, and generates a dataset with the necessary information organized.

[0259] Step 3:

[0260] The server uses pre-processed data to train a generative model. Here, natural language processing techniques are used to extract features from successful posts and learn patterns that lead to high engagement. These learning results are then incorporated into the generative model.

[0261] Step 4:

[0262] The server uses a pre-trained model to automatically generate new posts aligned with specific themes. This includes introductions to new products and information about specific campaigns. The generated content is then verified to ensure that its formatting and presentation are optimized.

[0263] Step 5:

[0264] The device receives instructions from the server and publishes the generated post to the appropriate social networking platform. The post is published at the optimal time based on the algorithm and user trends of the selected platform.

[0265] Step 6:

[0266] Users view posts on social media and react to them. Engagements such as comments and "likes" indicate the user's level of engagement with the post and influence subsequent server processing.

[0267] Step 7:

[0268] The server collects performance data for each post and analyzes engagement metrics (e.g., follower growth rate, likes, shares). This analysis is then used to improve future post creation and marketing strategies.

[0269] (Example 1)

[0270] Next, we will describe Example 1. 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."

[0271] The problem this invention aims to solve is to automate everything from data collection to content generation, publication, and user engagement optimization in order to conduct effective marketing activities on social media. In particular, it is necessary to process large amounts of data in real time and use natural language processing to filter and normalize it, thereby enabling the generation of effective content based on lean information. Furthermore, it is necessary to review and optimize the generated content and provide automated responses to maximize user engagement.

[0272] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0273] In this invention, the server includes information gathering means, preprocessing means, generative model learning means, information generation means, information publication means, communication means, analysis means, means for reviewing and optimizing the generated content, and means for providing automated responses to maximize user engagement. This enables efficient and effective marketing activities on social media, real-time data processing, optimization of information through natural language processing, and interaction with users.

[0274] An "information gathering device" is a device that has the function of acquiring information from multiple sources via a communication network and collecting data in real time.

[0275] A "preprocessing device" is a device that filters out unnecessary information from acquired data and then normalizes the data using natural language processing.

[0276] A "generative model learning device" is a device that uses pre-processed data to train a generative AI model and acquire patterns to maximize user engagement.

[0277] An "information generation means" is a device that automatically generates new information using a pre-trained generative model and creates content for posting.

[0278] The "information disclosure means" is a device that discloses the generated content to a designated online platform and posts it considering the optimal time zone and hashtags.

[0279] The "communication means" is a device for transmitting and receiving data and instructions in real time between the server and the terminal.

[0280] The "analysis means" is a device that analyzes data such as user reactions and changes in follows, and extracts information that can be used to improve the next marketing activity.

[0281] The "means for reviewing and optimizing the generated content" is a device that constructs an AI-based feedback loop and automatically improves the quality of the generated content.

[0282] The "means for performing an automatic response to maximize user engagement" is a device that automatically makes an appropriate response to a user's comment and deepens the interaction with the user.

[0283] The present invention is an advanced automation system for effectively implementing marketing activities on SNS by utilizing an AI agent. This system is composed of three main elements: a server, a terminal, and a user. The respective roles and specific usage methods are described below.

[0284] Role of the server

[0285] The server has a central role in performing information collection, processing, generation, and analysis. First, the server uses information collection means to obtain data from multiple information sources through a communication network. This data includes past posts, current trends, the trends of competing companies, and the like. Next, as preprocessing means, the server utilizes Pandas of Python and natural language processing tools to perform data filtering and normalization, and eliminate unnecessary information.

[0286] After that, the generation model learning means uses TensorFlow or PyTorch to learn the patterns of the data collected. In this learning, it explores the optimal method to enhance user engagement. This trained model is utilized as the information generation means to automatically generate content using a text generation model (e.g., GPT). Furthermore, the server reviews the generated content and improves the quality of the content by optimizing it through the AI feedback loop.

[0287] Role of the terminal

[0288] Based on the instructions sent from the server, the terminal is responsible for publishing the generated content to the SNS platform. The terminal receives data in real time using the WebSocket protocol and posts the content considering the optimal time zone and hashtags. This enables efficient reach to the targeted user layer.

[0289] Role of the user

[0290] The user browses the posts through the terminal and makes comments, "likes", shares, etc. This interaction is fed back to the server via the communication means and reflected in the next marketing activities by the analysis means.

[0291] Examples of specific cases and prompt texts

[0292] For example, in the promotion of a new product, the server analyzes the latest market trends and generates relevant content. This content is posted on SNS in a form that proposes the unique features and usage scenarios of the product. The user reacts to this content, and the data is utilized by the analysis means for the generation of the next content.

[0293] Examples of prompt texts include "Based on the latest market trends, generate an SNS post regarding the next new product. Maximize the engagement of followers."

[0294] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0295] Step 1:

[0296] The server retrieves data from multiple sources using information gathering tools. The input consists of data provided by each source. This data includes social media posts, trend information, and competitor activities. APIs are used to collect information in real time during data acquisition. The output is a collection of the retrieved raw data.

[0297] Step 2:

[0298] The server cleans the raw data obtained using preprocessing methods. The input is the raw data obtained in step 1. Specifically, the Pandas library in Python is used to filter out spam data and noise and normalize the data. Natural language processing tools are also used to format the text data. The output is refined, clean data.

[0299] Step 3:

[0300] The server uses a generative model training method to train a model on clean data. The input is the purified data obtained in step 2. This training is performed using machine learning libraries such as TensorFlow and PyTorch, and explores patterns that enhance user engagement. The output of this process is a trained generative AI model.

[0301] Step 4:

[0302] The server utilizes information generation means and uses a trained generation model to generate new content. The input is the trained model obtained in step 3 and the prompt text. As a specific operation, a text generation model (e.g., GPT) automatically generates content for posting based on the instructions in the prompt text. This output becomes the text data scheduled for posting.

[0303] Step 5:

[0304] The terminal publishes the content generated via the information disclosure means on the SNS platform. The input is the generated content obtained in step 4. The terminal transmits the content using the WebSocket protocol while considering an appropriate time zone and highly relevant hashtags. The output is the post published on the SNS.

[0305] Step 6:

[0306] The user views the published content and makes comments or reactions. The input is the post on the SNS. The user's interaction is fed back to the server and is utilized for the next data collection and content generation. The output is the user's reaction data.

[0307] Step 7:

[0308] The server utilizes analysis means to analyze the reaction data from the user and obtains insights for improving the next marketing strategy. The input is the user's reaction data obtained in step 6. In the analysis, indicators such as the increase or decrease in followers and the engagement rate are used to evaluate the effectiveness of the content. This output is a proposal for an improved marketing strategy.

[0309] (Application Example 1)

[0310] Next, Application Example 1 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0311] In today's advertising industry, there is a demand for personalized ads that are displayed at the right time based on users' interests and preferences. However, traditional advertising systems struggle to generate and display ads that respond to user behavior in real time, which limits the effectiveness of marketing activities.

[0312] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0313] In this invention, the server includes an information gathering device, a data processing device, and a generative model learning device. This makes it possible to acquire user behavior data in real time and generate and display advertising content optimized for the user based on that data.

[0314] An "information gathering device" is a device used to acquire data from individual information sources via an information network.

[0315] A "data processing device" is a device that analyzes collected data and transforms it into an optimized form based on user behavior and interests.

[0316] A "generative model learning device" is a device that uses processed data to build and train a learning model for generating advertising content based on user interests.

[0317] An "information generation device" is a device that has the function of actually generating advertisement copy and content using a trained model.

[0318] An "information transmission device" is a device that delivers generated advertising content to users through a user interface.

[0319] A "mutual communication device" is a device that transmits user responses and feedback to a server and uses them as data for generating the next content.

[0320] A "data analysis device" is a device that analyzes user response data and provides information for optimizing marketing strategies.

[0321] A "user interface device" is a device that allows users to directly interact with advertising content visually.

[0322] A "personalized display device" is a device that displays individually optimized advertisements to users according to their needs and interests.

[0323] To realize this invention, the system has the following configuration: The server operates with an information gathering device, a data processing device, and a generative model learning device. The information gathering device is responsible for collecting user data from social networking services (SNS) and other information sources via an information network. The data processing device cleanses the collected data and organizes it appropriately based on the user's interests and behavior. This process includes filtering spam data and removing unnecessary information.

[0324] Next, the generative model learning system uses the cleansed data and machine learning libraries such as TensorFlow to train a model for generating advertising content optimized for user interests. This enables the generation of new patterns based on past successful advertisements.

[0325] The terminal has a user interface device and is responsible for displaying generated advertising content to the user. When a user views an advertisement using the terminal, their response is sent to the server via a communication device. This response is analyzed in real time by the server's data analysis device and used to generate content for the next time.

[0326] For example, if a user frequently views sports-related posts, the generative model will use that data to create advertisements for the latest sports equipment. A possible prompt might be, "Based on the following information, create an ad copy that is likely to interest the user: The user's profile shows a lot of sports activity and they have liked many running-related posts." In this way, more appropriate and effective marketing activities become possible.

[0327] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0328] Step 1:

[0329] The server uses information gathering devices to collect user-related data from the information network. This data includes information about the user's activity history and interests on social networking services (SNS). It uses data obtained through SNS APIs as input and outputs it in a standardized format.

[0330] Step 2:

[0331] The server uses a data processing unit to cleanse the collected data. This process filters out spam and irrelevant information and extracts characteristic data based on user behavior. The input is the collected raw data, and the output is clean, analyzable data.

[0332] Step 3:

[0333] The server uses a generative model learning system to train a generative AI model using cleansed data and tools such as TensorFlow. In this step, patterns are identified to optimize user engagement. The input is the analyzed data obtained in step 2, and the output is the trained AI model.

[0334] Step 4:

[0335] Using the generated model, the server generates personalized advertising content tailored to the user's interests through an information generator. Here, ad copy is created using prompts. The input is a trained AI model and prompts, and the output is the specific ad content.

[0336] Step 5:

[0337] The terminal uses a user interface device to display generated advertising content to the user. This display is done via a smartphone or smart glasses. The input is advertising content received from the server, and the output is the advertisement visually presented to the user.

[0338] Step 6:

[0339] When a user interacts with an advertisement via their device, their response is sent to the server via a communication device. This response includes the number of clicks on the advertisement and comments. The input is user behavior data, and the output is feedback data for future content generation.

[0340] Step 7:

[0341] The server uses data analysis equipment to analyze user feedback and use it to improve advertising strategies. These analysis results are then incorporated into the creation of new advertising content. The input is feedback data, and the output is information for improving advertising strategies.

[0342] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0343] This invention incorporates an emotion engine that recognizes user emotions into a system that utilizes an AI agent to conduct effective marketing activities on social media. This allows for real-time analysis of the user's emotional state and optimization of marketing activities based on this analysis. The specific functions and usage examples for each element—server, terminal, and user—are described below.

[0344] Server Role

[0345] The server acts as the central hub of the system, handling data collection, processing, generation, and analysis. It also utilizes an emotion engine to analyze users' emotional states. Initially, the server collects posted data from social media platforms using various information gathering methods. By focusing on collecting data from particularly popular posts, comments, and reactions, it obtains a wealth of data regarding users' emotions.

[0346] Next, the data is cleaned using a preprocessing mechanism to organize it into a state suitable for training the generative model. Unnecessary data is removed during this process. The generative model training mechanism then performs model training incorporating sentiment data based on the preprocessed data. This enables content generation that takes into account changes in the user's emotions.

[0347] The emotional data acquired by the emotion engine helps generate appropriate content for users through information generation methods. For example, it can create encouraging messages for users who show positive reactions and supportive posts for users who show negative reactions.

[0348] Terminal role

[0349] The device publishes content generated from the server to the social networking platform at the appropriate time. This includes scheduling and adjusting posts based on sentiment analysis results. By posting during times when users are most active, engagement can be maximized.

[0350] User roles

[0351] Users interact with content provided through social media platforms. User comments and reactions are analyzed in real time using an emotion engine, and feedback is immediately provided by agents. Understanding each user's individual emotional state enables personalized responses.

[0352] Specific example

[0353] For example, during a campaign announcing a product launch, the server learns typical user response patterns from relevant historical data and continuously monitors users' real-time reactions using an emotion engine. In this case, it automatically generates posts offering coupons to users with positive emotions and provides special support guidance to users with negative emotions, thereby improving the overall user experience.

[0354] Thus, the present invention is a system that can realize more personalized marketing activities that take user emotions into consideration.

[0355] The following describes the processing flow.

[0356] Step 1:

[0357] The server collects post data related to predefined keywords and hashtags through the SNS platform's API. It prioritizes collecting posts that have garnered particularly high attention and stores this data in a database.

[0358] Step 2:

[0359] The server preprocesses the collected data. This process filters out unnecessary information and spam, extracting important data. It also cleans the text necessary for sentiment analysis and reduces noise.

[0360] Step 3:

[0361] The server trains a generative model using pre-processed data, while simultaneously using a sentiment engine to analyze sentiment data for each post and comment. The sentiment engine distinguishes between positive, negative, and neutral sentiments, tags them, and stores them.

[0362] Step 4:

[0363] Based on generative model training, the server automatically generates new posts that reflect sentiment data. Here, messages are adjusted according to the user's emotions, and content is constructed with an appropriate tone.

[0364] Step 5:

[0365] The system optimizes the timing of when the device receives content generated from the server and publishes it to the social networking platform. During this process, the posting schedule is adjusted based on the user's activity time and past posting patterns.

[0366] Step 6:

[0367] Users comment on and react to posts on social media, and their emotional states are recorded within the system. User feedback is reflected in real time and analyzed by the system's emotion engine.

[0368] Step 7:

[0369] The server analyzes user reaction data to posts and generates engagement metrics. This data is used to adjust future content creation and marketing strategies, optimizing the entire system.

[0370] (Example 2)

[0371] Next, we will describe Example 2. 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".

[0372] In today's digital society, effective marketing activities that take user emotions into account require the rapid and accurate collection of data from diverse sources and the selection of useful information from that data. However, traditional methods are inefficient in data collection and analysis, making it difficult to grasp user emotions in real time and reflect them in marketing strategies.

[0373] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0374] In this invention, the server includes an information acquisition means, a numerical processing means, and an emotion analysis means. This makes it possible to quickly collect necessary information from a large amount of data on an SNS platform, effectively remove unnecessary data, and then accurately analyze the user's emotions. This enables personalized marketing activities that respond to changes in the user's emotions.

[0375] "Information acquisition means" refers to a mechanism for obtaining necessary data from multiple information sources using wide-area communication.

[0376] "Numerical processing means" refers to functions that remove noise and unnecessary data from acquired data and prepare it in a format suitable for analysis.

[0377] "Computational model learning methods" refer to the process of training machine learning models using pre-processed data to obtain insights that are useful for future predictions and generation.

[0378] "Information creation means" refers to methods for generating new information or content based on a learned model.

[0379] "Information distribution means" refers to the function of publishing generated information to the appropriate platform at the optimal time.

[0380] "Information transmission means" refers to communication circuits and protocols used to exchange data within and outside a system.

[0381] "Evaluation methods" refer to methods for analyzing the performance of generated information and the overall system to identify areas for improvement.

[0382] "Emotional analysis tools" are functions that extract the user's emotional state from data and classify and analyze it.

[0383] This invention is a system based on advanced information technology that analyzes user emotions on a social networking service (SNS) platform and enables personalized marketing based on that analysis. The system is composed of an information acquisition means, a numerical processing means, an emotion analysis means, a computational model learning means, an information creation means, an information distribution means, and an information transmission means.

[0384] The server first acquires a large amount of user data from SNS platforms and other sources via wide-area communication using information acquisition methods. Specifically, it collects data primarily from posts, comments, and reactions that attract particularly high attention through API access. Based on this data, the server performs preprocessing using numerical processing methods to remove noise and unnecessary data, preparing it for analysis. Open-source libraries (e.g., NLTK and spaCy) can be used as natural language processing tools.

[0385] Next, the server analyzes the sentiment of the pre-processed data using sentiment analysis tools. Existing natural language processing APIs (e.g., Google Cloud Natural Language API) are used for sentiment analysis, classifying the user's emotional state with tags such as positive, negative, and neutral. Based on these analysis results, a generative AI model is trained using computational model learning tools. TensorFlow and PyTorch are expected to be used as machine learning frameworks.

[0386] The server uses the acquired emotion data to generate content that responds to the user's emotions using an information creation mechanism. For example, it is possible to input a prompt such as, "Generate an encouraging message based on the user's positive response data," and the AI ​​can dynamically generate a message.

[0387] The generated content is published to social media platforms via information distribution methods at the optimal time determined by the device. The device analyzes user activity patterns and adjusts posts to coincide with the time when engagement is maximized. This can enhance the overall effectiveness of marketing activities.

[0388] Users react to and comment on the delivered content. This feedback is sent back to the server via communication channels and used for sentiment analysis and content generation in the next round. The learning process of the generative AI model is continuously improved by incorporating user feedback, enabling the delivery of more effective marketing messages.

[0389] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0390] Step 1:

[0391] The server uses information acquisition methods to obtain large amounts of user data from SNS platforms via wide-area communication. The input is raw posted data via API, and the output is a collection of raw data organized in chronological order. At this stage, priority is given to collecting posts with a large number of "likes" and "shares."

[0392] Step 2:

[0393] The server performs preprocessing on the acquired raw data using numerical processing tools. The input is the raw data obtained in step 1, and the output is the cleaned-up data. This preprocessing includes removing noise data and spam posts, and tokenizing comments and posts. The text is morphologically analyzed using a natural language processing library and formatted into a format suitable for analysis.

[0394] Step 3:

[0395] The server uses sentiment analysis tools to analyze pre-processed data and classify the user's emotional state. The input is cleaned-up data, and the output is data tagged with sentiment (positive, negative, neutral). A dedicated API is used for sentiment analysis, and accuracy is improved by scoring the intensity of each emotion.

[0396] Step 4:

[0397] The server uses computational model learning methods to train a generative AI model based on emotion-tagged data. The input is emotion-tagged data, and the output is a model that can predict changes in the user's emotions. A machine learning framework is used to extract features from the data and create an optimized model.

[0398] Step 5:

[0399] The server generates content using a trained model and prompt text via an information creation mechanism. The input is the trained model and the prompt text "Generate an encouraging message based on the user's positive response data," and the output is a personalized message. The AI ​​suggests the most appropriate wording based on the user's past response patterns.

[0400] Step 6:

[0401] The device publishes generated content to social media platforms at the optimal time via an information distribution method. The input is the generated content, and the output is a successful posting message to social media. The device analyzes past user activity times and delivers content at the time when engagement is highest.

[0402] Step 7:

[0403] Users provide feedback by reacting to and commenting on published content. The input is the content published on social media, and the output is new comments and reactions from users. This feedback is sent back to the server and used for sentiment analysis and content generation in the future.

[0404] (Application Example 2)

[0405] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0406] In marketing activities on social media, there is a challenge in that personalized ad delivery that takes into account the individual emotions of users is not being adequately implemented. As a result, ads that do not match users' interests or emotions are often delivered, which can lead to decreased ad effectiveness and a decline in the quality of the user experience.

[0407] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0408] In this invention, the server includes information gathering means, sentiment analysis means, and advertisement generation means. This makes it possible to analyze the user's emotions based on data collected from the SNS platform and deliver advertisements that are appropriate to the user's emotions in real time based on the analysis results.

[0409] "Information gathering means" refers to a system that has the function of acquiring information from multiple sources via a communication network.

[0410] A "preprocessing means" is a device that has the function of filtering out unnecessary information from the acquired information and using the refined information.

[0411] A "generative model learning tool" is a device that has the function of enabling an AI model to learn based on pre-processed data.

[0412] An "information generation means" is a device that has the function of generating appropriate content based on user sentiment data.

[0413] "Information disclosure means" refers to a system that has the functionality to publish generated content on a social networking service (SNS) platform.

[0414] "Communication means" refers to anything that has the function of sending and receiving data between each element within a system.

[0415] "Analysis tools" are those that analyze collected data and have the function of revealing users' emotions and behavioral patterns.

[0416] An "emotion analysis tool" is a device that has the function of analyzing the user's emotions from the collected data.

[0417] An "advertising generation method" is a device that has the function of generating advertisements appropriate to the user based on the results of sentiment analysis.

[0418] The server uses information gathering means to collect posted data from SNS platforms. In this process, it obtains the latest user posts and reaction data from Twitter and Facebook APIs via a communication network. This collected data is then filtered and refined by preprocessing means. Based on this refined data, a generative model learning means performs user sentiment analysis using an AI model (e.g., BERT or GPT).

[0419] The results of the sentiment analysis are used by the ad generation system to generate ads that are appropriate to the user's emotions. These ads are then prepared for posting to social media platforms via the information generation system and published at the appropriate time by the information publication system.

[0420] The device publishes this advertisement on social media based on instructions from the server and displays it at the appropriate time to increase user engagement. Users react to the published advertisement, which collects even more sentiment data that is used as training data for the entire system.

[0421] For example, if a user posts "Today is the best day!" on social media, the server analyzes this as a positive emotion and generates a restaurant coupon for a new product using an ad generation tool. Ultimately, this ad will appear in the user's feed.

[0422] Example prompt: "For posts that show positive user emotions, display ads that enhance enjoyment; for posts that show negative emotions, display ads that soothe the soul."

[0423] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0424] Step 1:

[0425] The server uses information gathering tools to obtain posted data and reaction data from the SNS platform's API via the communication network. In this process, messages and comments generated by users on SNS are collected as input, and the data is output in a standard data format for use in subsequent data processing.

[0426] Step 2:

[0427] The server uses preprocessing to filter out unnecessary data from the data collected in step 1 and extract the necessary information. In this process, since the input data may contain noise or duplicate data, these are removed, and clean data in a format suitable for sentiment analysis is output.

[0428] Step 3:

[0429] The server passes the refined data to a generative model training tool, which then uses an AI model (e.g., BERT or GPT) to analyze the user's emotions. This data calculation outputs a user's emotion score (positive, negative, neutral) based on the input clean data.

[0430] Step 4:

[0431] The server receives the output from the sentiment analysis system and uses the ad generation system to create ads that match the user's emotions. Here, the user's sentiment score is used as input, and highly relevant ad campaigns and promotional information are output.

[0432] Step 5:

[0433] The terminal uses information generation means to prepare to post the advertisement generated in step 4 to the SNS platform. In this preparation stage, the input advertisement data is converted into an appropriate format and output with scheduling information added.

[0434] Step 6:

[0435] The device publishes advertisements generated at the appropriate time on social media through an information disclosure mechanism. Here, posts prepared using the social media API are used as input data, timely publication processing is performed, and the result of the advertisement appearing in the user's feed is output.

[0436] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0437] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0438] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0439] [Third Embodiment]

[0440] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0441] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0442] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0443] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0444] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0445] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0446] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0447] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0448] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0449] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0450] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0451] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0452] This invention is a system for conducting effective marketing activities on social media using an AI agent. This system consists of three main elements: a server, a terminal, and a user. The processing of each element and specific examples are shown below.

[0453] Server Role

[0454] The server plays a central role in collecting, processing, generating, and analyzing information. First, the server collects data from designated online platforms. This includes data related to past successful posts and current trends. Next, the data acquired through information gathering is preprocessed to prepare it for training generative models. At this stage, it is crucial to remove spam data and noise to complete the cleaning process.

[0455] Subsequently, the generative model training system performs model training using pre-processed data. This generative model analyzes past successful cases and acquires patterns to maximize user engagement. After training is complete, the information generation system utilizes this model to automatically generate effective posts. The content thus generated is then reviewed and optimized by the server.

[0456] Terminal role

[0457] The terminal's primary role is to publish generated content on selected social media platforms based on instructions sent from the server. The published posts utilize content created by a generation AI, taking into account factors such as time of day and hashtags. This optimizes reach to the target user base.

[0458] User roles

[0459] Users interact with the information and content provided by this system. Engagement can be increased by having agents automatically respond to user comments via communication channels. Data such as user responses and changes in follower counts are sent to the server via analytical tools and used for future marketing activities.

[0460] Specific example

[0461] For example, when promoting a new product, the server analyzes the latest market trends and generates highly relevant content. This content, such as "the new product's distinctive features" and "suggested usage scenarios," is posted to social media via the user's device. Users view these posts, comment, and react, which are immediately analyzed by the server and reflected in the generation of subsequent content. This process allows companies to quickly and cost-effectively increase product awareness and promote sales.

[0462] In the above configuration, the present invention functions as a system that can efficiently acquire followers on social media and implement effective marketing activities.

[0463] The following describes the processing flow.

[0464] Step 1:

[0465] The server uses the SNS platform's API to collect relevant post data based on specific keywords and hashtags. A filter is applied to list posts that have previously received high engagement.

[0466] Step 2:

[0467] The server preprocesses the collected data. This process removes duplicate and spam data, cleans the text, filters out irrelevant noise, and generates a dataset with the necessary information organized.

[0468] Step 3:

[0469] The server uses pre-processed data to train a generative model. Here, natural language processing techniques are used to extract features from successful posts and learn patterns that lead to high engagement. These learning results are then incorporated into the generative model.

[0470] Step 4:

[0471] The server uses a pre-trained model to automatically generate new posts aligned with specific themes. This includes introductions to new products and information about specific campaigns. The generated content is then verified to ensure that its formatting and presentation are optimized.

[0472] Step 5:

[0473] The device receives instructions from the server and publishes the generated post to the appropriate social networking platform. The post is published at the optimal time based on the algorithm and user trends of the selected platform.

[0474] Step 6:

[0475] Users view posts on social media and react to them. Engagements such as comments and "likes" indicate the user's level of engagement with the post and influence subsequent server processing.

[0476] Step 7:

[0477] The server collects performance data for each post and analyzes engagement metrics (e.g., follower growth rate, likes, shares). This analysis is then used to improve future post creation and marketing strategies.

[0478] (Example 1)

[0479] Next, we will describe Example 1. 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."

[0480] The problem this invention aims to solve is to automate everything from data collection to content generation, publication, and user engagement optimization in order to conduct effective marketing activities on social media. In particular, it is necessary to process large amounts of data in real time and use natural language processing to filter and normalize it, thereby enabling the generation of effective content based on lean information. Furthermore, it is necessary to review and optimize the generated content and provide automated responses to maximize user engagement.

[0481] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0482] In this invention, the server includes information gathering means, preprocessing means, generative model learning means, information generation means, information publication means, communication means, analysis means, means for reviewing and optimizing the generated content, and means for providing automated responses to maximize user engagement. This enables efficient and effective marketing activities on social media, real-time data processing, optimization of information through natural language processing, and interaction with users.

[0483] An "information gathering device" is a device that has the function of acquiring information from multiple sources via a communication network and collecting data in real time.

[0484] A "preprocessing device" is a device that filters out unnecessary information from acquired data and then normalizes the data using natural language processing.

[0485] A "generative model learning device" is a device that uses pre-processed data to train a generative AI model and acquire patterns to maximize user engagement.

[0486] An "information generation means" is a device that automatically generates new information using a pre-trained generative model and creates content for posting.

[0487] "Information disclosure means" refers to a device that publishes generated content on a designated online platform and posts it considering the optimal time and hashtags.

[0488] A "communication device" is a device used to send and receive data and instructions in real time between a server and a terminal.

[0489] An "analysis tool" is a device that analyzes data such as user responses and changes in follow-up activity to extract information that can be used to improve future marketing activities.

[0490] "Means for reviewing and optimizing generated content" refers to a system that builds an AI-driven feedback loop to automatically improve the quality of generated content.

[0491] "Means of providing automated responses to maximize user engagement" refers to a device that automatically provides appropriate responses to user comments, thereby deepening interaction with the user.

[0492] This invention is an advanced automated system for effectively conducting marketing activities on social media using an AI agent. This system consists of three main elements: a server, a terminal, and a user. The roles and specific usage methods of each are described below.

[0493] Server Role

[0494] The server plays a central role in everything from information gathering to processing, generation, and analysis. First, the server uses information gathering tools to acquire data from multiple sources via the communication network. This data includes past posts, current trends, and the activities of competitors. Next, the server uses Python's Pandas and natural language processing tools as preprocessing tools to filter and normalize the data and remove unnecessary information.

[0495] Subsequently, the system learns patterns in the collected data using generative model training methods such as TensorFlow or PyTorch. This training explores the optimal methods for increasing user engagement. This trained model is then used as an information generation tool, automatically generating content using a text generation model (e.g., GPT). Furthermore, the server reviews the generated content and improves its quality through an AI feedback loop for optimization.

[0496] Terminal role

[0497] The terminal is responsible for publishing generated content to the SNS platform based on instructions sent from the server. The terminal receives data in real time using the WebSocket protocol and posts content considering the optimal time and hashtags. This allows for efficient reach to the target user base.

[0498] User roles

[0499] Users view posts through their devices and can comment, like, and share them. This interaction is fed back to the server via communication methods and analyzed to inform future marketing activities.

[0500] Examples of specific cases and prompt statements

[0501] For example, in the promotion of a new product, a server analyzes the latest market trends and generates highly relevant content. This content is posted on social media in a format that suggests the product's unique features and usage scenarios. Users react to this content, and this data is analyzed and used to generate subsequent content.

[0502] An example of a prompt might be: "Generate a social media post about the next new product based on the latest market trends. Maximize follower engagement."

[0503] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0504] Step 1:

[0505] The server retrieves data from multiple sources using information gathering tools. The input consists of data provided by each source. This data includes social media posts, trend information, and competitor activities. APIs are used to collect information in real time during data acquisition. The output is a collection of the retrieved raw data.

[0506] Step 2:

[0507] The server cleans the raw data obtained using preprocessing methods. The input is the raw data obtained in step 1. Specifically, the Pandas library in Python is used to filter out spam data and noise and normalize the data. Natural language processing tools are also used to format the text data. The output is refined, clean data.

[0508] Step 3:

[0509] The server uses a generative model training method to train a model on clean data. The input is the purified data obtained in step 2. This training is performed using machine learning libraries such as TensorFlow and PyTorch, and explores patterns that enhance user engagement. The output of this process is a trained generative AI model.

[0510] Step 4:

[0511] The server utilizes information generation tools and generates new content using a trained generative model. The input consists of the trained model and prompt text obtained in step 3. Specifically, a text generation model (e.g., GPT) automatically generates content for posting based on the instructions in the prompt text. This output becomes the text data to be posted.

[0512] Step 5:

[0513] The device publishes the content generated via the information disclosure mechanism on the SNS platform. The input is the generated content obtained in step 4. The device sends the content using the WebSocket protocol, taking into account the appropriate time of day and relevant hashtags. The output is the post published on SNS.

[0514] Step 6:

[0515] Users view published content and leave comments and reactions. Input is posts on social media. User interactions are fed back to the server, which is used for future data collection and content generation. Output is user response data.

[0516] Step 7:

[0517] The server uses analytical tools to analyze user response data and gain insights to improve future marketing strategies. The input is the user response data obtained in step 6. The analysis evaluates the effectiveness of the content using metrics such as follower growth / decrease and engagement rate. The output is a proposal for an improved marketing strategy.

[0518] (Application Example 1)

[0519] Next, we will explain Application Example 1. In the following explanation, 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."

[0520] In today's advertising industry, there is a demand for personalized ads that are displayed at the right time based on users' interests and preferences. However, traditional advertising systems struggle to generate and display ads that respond to user behavior in real time, which limits the effectiveness of marketing activities.

[0521] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0522] In this invention, the server includes an information gathering device, a data processing device, and a generative model learning device. This makes it possible to acquire user behavior data in real time and generate and display advertising content optimized for the user based on that data.

[0523] An "information gathering device" is a device used to acquire data from individual information sources via an information network.

[0524] A "data processing device" is a device that analyzes collected data and transforms it into an optimized form based on user behavior and interests.

[0525] A "generative model learning device" is a device that uses processed data to build and train a learning model for generating advertising content based on user interests.

[0526] An "information generation device" is a device that has the function of actually generating advertisement copy and content using a trained model.

[0527] An "information transmission device" is a device that delivers generated advertising content to users through a user interface.

[0528] A "mutual communication device" is a device that transmits user responses and feedback to a server and uses them as data for generating the next content.

[0529] A "data analysis device" is a device that analyzes user response data and provides information for optimizing marketing strategies.

[0530] A "user interface device" is a device that allows users to directly interact with advertising content visually.

[0531] A "personalized display device" is a device that displays individually optimized advertisements to users according to their needs and interests.

[0532] To realize this invention, the system has the following configuration: The server operates with an information gathering device, a data processing device, and a generative model learning device. The information gathering device is responsible for collecting user data from social networking services (SNS) and other information sources via an information network. The data processing device cleanses the collected data and organizes it appropriately based on the user's interests and behavior. This process includes filtering spam data and removing unnecessary information.

[0533] Next, the generative model learning system uses the cleansed data and machine learning libraries such as TensorFlow to train a model for generating advertising content optimized for user interests. This enables the generation of new patterns based on past successful advertisements.

[0534] The terminal has a user interface device and is responsible for displaying generated advertising content to the user. When a user views an advertisement using the terminal, their response is sent to the server via a communication device. This response is analyzed in real time by the server's data analysis device and used to generate content for the next time.

[0535] For example, if a user frequently views sports-related posts, the generative model will use that data to create advertisements for the latest sports equipment. A possible prompt might be, "Based on the following information, create an ad copy that is likely to interest the user: The user's profile shows a lot of sports activity and they have liked many running-related posts." In this way, more appropriate and effective marketing activities become possible.

[0536] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0537] Step 1:

[0538] The server uses information gathering devices to collect user-related data from the information network. This data includes information about the user's activity history and interests on social networking services (SNS). It uses data obtained through SNS APIs as input and outputs it in a standardized format.

[0539] Step 2:

[0540] The server uses a data processing unit to cleanse the collected data. This process filters out spam and irrelevant information and extracts characteristic data based on user behavior. The input is the collected raw data, and the output is clean, analyzable data.

[0541] Step 3:

[0542] The server uses a generative model learning system to train a generative AI model using cleansed data and tools such as TensorFlow. In this step, patterns are identified to optimize user engagement. The input is the analyzed data obtained in step 2, and the output is the trained AI model.

[0543] Step 4:

[0544] Using the generated model, the server generates personalized advertising content tailored to the user's interests through an information generator. Here, ad copy is created using prompts. The input is a trained AI model and prompts, and the output is the specific ad content.

[0545] Step 5:

[0546] The terminal uses a user interface device to display generated advertising content to the user. This display is done via a smartphone or smart glasses. The input is advertising content received from the server, and the output is the advertisement visually presented to the user.

[0547] Step 6:

[0548] When a user interacts with an advertisement via their device, their response is sent to the server via a communication device. This response includes the number of clicks on the advertisement and comments. The input is user behavior data, and the output is feedback data for future content generation.

[0549] Step 7:

[0550] The server uses data analysis equipment to analyze user feedback and use it to improve advertising strategies. These analysis results are then incorporated into the creation of new advertising content. The input is feedback data, and the output is information for improving advertising strategies.

[0551] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0552] This invention incorporates an emotion engine that recognizes user emotions into a system that utilizes an AI agent to conduct effective marketing activities on social media. This allows for real-time analysis of the user's emotional state and optimization of marketing activities based on this analysis. The specific functions and usage examples for each element—server, terminal, and user—are described below.

[0553] Server Role

[0554] The server acts as the central hub of the system, handling data collection, processing, generation, and analysis. It also utilizes an emotion engine to analyze users' emotional states. Initially, the server collects posted data from social media platforms using various information gathering methods. By focusing on collecting data from particularly popular posts, comments, and reactions, it obtains a wealth of data regarding users' emotions.

[0555] Next, the data is cleaned using a preprocessing mechanism to organize it into a state suitable for training the generative model. Unnecessary data is removed during this process. The generative model training mechanism then performs model training incorporating sentiment data based on the preprocessed data. This enables content generation that takes into account changes in the user's emotions.

[0556] The emotional data acquired by the emotion engine helps generate appropriate content for users through information generation methods. For example, it can create encouraging messages for users who show positive reactions and supportive posts for users who show negative reactions.

[0557] Terminal role

[0558] The device publishes content generated from the server to the social networking platform at the appropriate time. This includes scheduling and adjusting posts based on sentiment analysis results. By posting during times when users are most active, engagement can be maximized.

[0559] User roles

[0560] Users interact with content provided through social media platforms. User comments and reactions are analyzed in real time using an emotion engine, and feedback is immediately provided by agents. Understanding each user's individual emotional state enables personalized responses.

[0561] Specific example

[0562] For example, during a campaign announcing a product launch, the server learns typical user response patterns from relevant historical data and continuously monitors users' real-time reactions using an emotion engine. In this case, it automatically generates posts offering coupons to users with positive emotions and provides special support guidance to users with negative emotions, thereby improving the overall user experience.

[0563] Thus, the present invention is a system that can realize more personalized marketing activities that take user emotions into consideration.

[0564] The following describes the processing flow.

[0565] Step 1:

[0566] The server collects post data related to predefined keywords and hashtags through the SNS platform's API. It prioritizes collecting posts that have garnered particularly high attention and stores this data in a database.

[0567] Step 2:

[0568] The server preprocesses the collected data. This process filters out unnecessary information and spam, extracting important data. It also cleans the text necessary for sentiment analysis and reduces noise.

[0569] Step 3:

[0570] The server trains a generative model using pre-processed data, while simultaneously using a sentiment engine to analyze sentiment data for each post and comment. The sentiment engine distinguishes between positive, negative, and neutral sentiments, tags them, and stores them.

[0571] Step 4:

[0572] Based on generative model training, the server automatically generates new posts that reflect sentiment data. Here, messages are adjusted according to the user's emotions, and content is constructed with an appropriate tone.

[0573] Step 5:

[0574] The system optimizes the timing of when the device receives content generated from the server and publishes it to the social networking platform. During this process, the posting schedule is adjusted based on the user's activity time and past posting patterns.

[0575] Step 6:

[0576] Users comment on and react to posts on social media, and their emotional states are recorded within the system. User feedback is reflected in real time and analyzed by the system's emotion engine.

[0577] Step 7:

[0578] The server analyzes user reaction data to posts and generates engagement metrics. This data is used to adjust future content creation and marketing strategies, optimizing the entire system.

[0579] (Example 2)

[0580] Next, we will describe Example 2. 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."

[0581] In today's digital society, effective marketing activities that take user emotions into account require the rapid and accurate collection of data from diverse sources and the selection of useful information from that data. However, traditional methods are inefficient in data collection and analysis, making it difficult to grasp user emotions in real time and reflect them in marketing strategies.

[0582] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0583] In this invention, the server includes an information acquisition means, a numerical processing means, and an emotion analysis means. This makes it possible to quickly collect necessary information from a large amount of data on an SNS platform, effectively remove unnecessary data, and then accurately analyze the user's emotions. This enables personalized marketing activities that respond to changes in the user's emotions.

[0584] "Information acquisition means" refers to a mechanism for obtaining necessary data from multiple information sources using wide-area communication.

[0585] "Numerical processing means" refers to functions that remove noise and unnecessary data from acquired data and prepare it in a format suitable for analysis.

[0586] "Computational model learning methods" refer to the process of training machine learning models using pre-processed data to obtain insights that are useful for future predictions and generation.

[0587] "Information creation means" refers to methods for generating new information or content based on a learned model.

[0588] "Information distribution means" refers to the function of publishing generated information to the appropriate platform at the optimal time.

[0589] "Information transmission means" refers to communication circuits and protocols used to exchange data within and outside a system.

[0590] "Evaluation methods" refer to methods for analyzing the performance of generated information and the overall system to identify areas for improvement.

[0591] "Emotional analysis tools" are functions that extract the user's emotional state from data and classify and analyze it.

[0592] This invention is a system based on advanced information technology that analyzes user emotions on a social networking service (SNS) platform and enables personalized marketing based on that analysis. The system is composed of an information acquisition means, a numerical processing means, an emotion analysis means, a computational model learning means, an information creation means, an information distribution means, and an information transmission means.

[0593] The server first acquires a large amount of user data from SNS platforms and other sources via wide-area communication using information acquisition methods. Specifically, it collects data primarily from posts, comments, and reactions that attract particularly high attention through API access. Based on this data, the server performs preprocessing using numerical processing methods to remove noise and unnecessary data, preparing it for analysis. Open-source libraries (e.g., NLTK and spaCy) can be used as natural language processing tools.

[0594] Next, the server analyzes the sentiment of the pre-processed data using sentiment analysis tools. Existing natural language processing APIs (e.g., Google Cloud Natural Language API) are used for sentiment analysis, classifying the user's emotional state with tags such as positive, negative, and neutral. Based on these analysis results, a generative AI model is trained using computational model learning tools. TensorFlow and PyTorch are expected to be used as machine learning frameworks.

[0595] The server uses the acquired emotion data to generate content that responds to the user's emotions using an information creation mechanism. For example, it is possible to input a prompt such as, "Generate an encouraging message based on the user's positive response data," and the AI ​​can dynamically generate a message.

[0596] The generated content is published to social media platforms via information distribution methods at the optimal time determined by the device. The device analyzes user activity patterns and adjusts posts to coincide with the time when engagement is maximized. This can enhance the overall effectiveness of marketing activities.

[0597] Users react to and comment on the delivered content. This feedback is sent back to the server via communication channels and used for sentiment analysis and content generation in the next round. The learning process of the generative AI model is continuously improved by incorporating user feedback, enabling the delivery of more effective marketing messages.

[0598] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0599] Step 1:

[0600] The server uses information acquisition methods to obtain large amounts of user data from SNS platforms via wide-area communication. The input is raw posted data via API, and the output is a collection of raw data organized in chronological order. At this stage, priority is given to collecting posts with a large number of "likes" and "shares."

[0601] Step 2:

[0602] The server performs preprocessing on the acquired raw data using numerical processing tools. The input is the raw data obtained in step 1, and the output is the cleaned-up data. This preprocessing includes removing noise data and spam posts, and tokenizing comments and posts. The text is morphologically analyzed using a natural language processing library and formatted into a format suitable for analysis.

[0603] Step 3:

[0604] The server uses sentiment analysis tools to analyze pre-processed data and classify the user's emotional state. The input is cleaned-up data, and the output is data tagged with sentiment (positive, negative, neutral). A dedicated API is used for sentiment analysis, and accuracy is improved by scoring the intensity of each emotion.

[0605] Step 4:

[0606] The server uses computational model learning methods to train a generative AI model based on emotion-tagged data. The input is emotion-tagged data, and the output is a model that can predict changes in the user's emotions. A machine learning framework is used to extract features from the data and create an optimized model.

[0607] Step 5:

[0608] The server generates content using a trained model and prompt text via an information creation mechanism. The input is the trained model and the prompt text "Generate an encouraging message based on the user's positive response data," and the output is a personalized message. The AI ​​suggests the most appropriate wording based on the user's past response patterns.

[0609] Step 6:

[0610] The device publishes generated content to social media platforms at the optimal time via an information distribution method. The input is the generated content, and the output is a successful posting message to social media. The device analyzes past user activity times and delivers content at the time when engagement is highest.

[0611] Step 7:

[0612] Users provide feedback by reacting to and commenting on published content. The input is the content published on social media, and the output is new comments and reactions from users. This feedback is sent back to the server and used for sentiment analysis and content generation in the future.

[0613] (Application Example 2)

[0614] Next, we will explain Application Example 2. In the following explanation, 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."

[0615] In marketing activities on social media, there is a challenge in that personalized ad delivery that takes into account the individual emotions of users is not being adequately implemented. As a result, ads that do not match users' interests or emotions are often delivered, which can lead to decreased ad effectiveness and a decline in the quality of the user experience.

[0616] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0617] In this invention, the server includes information gathering means, sentiment analysis means, and advertisement generation means. This makes it possible to analyze the user's emotions based on data collected from the SNS platform and deliver advertisements that are appropriate to the user's emotions in real time based on the analysis results.

[0618] "Information gathering means" refers to a system that has the function of acquiring information from multiple sources via a communication network.

[0619] A "preprocessing means" is a device that has the function of filtering out unnecessary information from the acquired information and using the refined information.

[0620] A "generative model learning tool" is a device that has the function of enabling an AI model to learn based on pre-processed data.

[0621] An "information generation means" is a device that has the function of generating appropriate content based on user sentiment data.

[0622] "Information disclosure means" refers to a system that has the functionality to publish generated content on a social networking service (SNS) platform.

[0623] "Communication means" refers to anything that has the function of sending and receiving data between each element within a system.

[0624] "Analysis tools" are those that analyze collected data and have the function of revealing users' emotions and behavioral patterns.

[0625] An "emotion analysis tool" is a device that has the function of analyzing the user's emotions from the collected data.

[0626] An "advertising generation method" is a device that has the function of generating advertisements appropriate to the user based on the results of sentiment analysis.

[0627] The server uses information gathering means to collect posted data from SNS platforms. In this process, it obtains the latest user posts and reaction data from Twitter and Facebook APIs via a communication network. This collected data is then filtered and refined by preprocessing means. Based on this refined data, a generative model learning means performs user sentiment analysis using an AI model (e.g., BERT or GPT).

[0628] The results of the sentiment analysis are used by the ad generation system to generate ads that are appropriate to the user's emotions. These ads are then prepared for posting to social media platforms via the information generation system and published at the appropriate time by the information publication system.

[0629] The device publishes this advertisement on social media based on instructions from the server and displays it at the appropriate time to increase user engagement. Users react to the published advertisement, which collects even more sentiment data that is used as training data for the entire system.

[0630] For example, if a user posts "Today is the best day!" on social media, the server analyzes this as a positive emotion and generates a restaurant coupon for a new product using an ad generation tool. Ultimately, this ad will appear in the user's feed.

[0631] Example prompt: "For posts that show positive user emotions, display ads that enhance enjoyment; for posts that show negative emotions, display ads that soothe the soul."

[0632] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0633] Step 1:

[0634] The server uses information gathering tools to obtain posted data and reaction data from the SNS platform's API via the communication network. In this process, messages and comments generated by users on SNS are collected as input, and the data is output in a standard data format for use in subsequent data processing.

[0635] Step 2:

[0636] The server uses preprocessing to filter out unnecessary data from the data collected in step 1 and extract the necessary information. In this process, since the input data may contain noise or duplicate data, these are removed, and clean data in a format suitable for sentiment analysis is output.

[0637] Step 3:

[0638] The server passes the refined data to a generative model training tool, which then uses an AI model (e.g., BERT or GPT) to analyze the user's emotions. This data calculation outputs a user's emotion score (positive, negative, neutral) based on the input clean data.

[0639] Step 4:

[0640] The server receives the output from the sentiment analysis system and uses the ad generation system to create ads that match the user's emotions. Here, the user's sentiment score is used as input, and highly relevant ad campaigns and promotional information are output.

[0641] Step 5:

[0642] The terminal uses information generation means to prepare to post the advertisement generated in step 4 to the SNS platform. In this preparation stage, the input advertisement data is converted into an appropriate format and output with scheduling information added.

[0643] Step 6:

[0644] The device publishes advertisements generated at the appropriate time on social media through an information disclosure mechanism. Here, posts prepared using the social media API are used as input data, timely publication processing is performed, and the result of the advertisement appearing in the user's feed is output.

[0645] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0646] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0647] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0648] [Fourth Embodiment]

[0649] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0650] As shown in Figure 7, the 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.

[0651] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0652] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0653] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0654] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0655] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0656] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0657] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0658] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0659] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0660] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0661] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0662] This invention is a system for conducting effective marketing activities on social media using an AI agent. This system consists of three main elements: a server, a terminal, and a user. The processing of each element and specific examples are shown below.

[0663] Server Role

[0664] The server plays a central role in collecting, processing, generating, and analyzing information. First, the server collects data from designated online platforms. This includes data related to past successful posts and current trends. Next, the data acquired through information gathering is preprocessed to prepare it for training generative models. At this stage, it is crucial to remove spam data and noise to complete the cleaning process.

[0665] Subsequently, the generative model training system performs model training using pre-processed data. This generative model analyzes past successful cases and acquires patterns to maximize user engagement. After training is complete, the information generation system utilizes this model to automatically generate effective posts. The content thus generated is then reviewed and optimized by the server.

[0666] Terminal role

[0667] The terminal's primary role is to publish generated content on selected social media platforms based on instructions sent from the server. The published posts utilize content created by a generation AI, taking into account factors such as time of day and hashtags. This optimizes reach to the target user base.

[0668] User roles

[0669] Users interact with the information and content provided by this system. Engagement can be increased by having agents automatically respond to user comments via communication channels. Data such as user responses and changes in follower counts are sent to the server via analytical tools and used for future marketing activities.

[0670] Specific example

[0671] For example, when promoting a new product, the server analyzes the latest market trends and generates highly relevant content. This content, such as "the new product's distinctive features" and "suggested usage scenarios," is posted to social media via the user's device. Users view these posts, comment, and react, which are immediately analyzed by the server and reflected in the generation of subsequent content. This process allows companies to quickly and cost-effectively increase product awareness and promote sales.

[0672] In the above configuration, the present invention functions as a system that can efficiently acquire followers on social media and implement effective marketing activities.

[0673] The following describes the processing flow.

[0674] Step 1:

[0675] The server uses the SNS platform's API to collect relevant post data based on specific keywords and hashtags. A filter is applied to list posts that have previously received high engagement.

[0676] Step 2:

[0677] The server preprocesses the collected data. This process removes duplicate and spam data, cleans the text, filters out irrelevant noise, and generates a dataset with the necessary information organized.

[0678] Step 3:

[0679] The server uses pre-processed data to train a generative model. Here, natural language processing techniques are used to extract features from successful posts and learn patterns that lead to high engagement. These learning results are then incorporated into the generative model.

[0680] Step 4:

[0681] The server uses a pre-trained model to automatically generate new posts aligned with specific themes. This includes introductions to new products and information about specific campaigns. The generated content is then verified to ensure that its formatting and presentation are optimized.

[0682] Step 5:

[0683] The device receives instructions from the server and publishes the generated post to the appropriate social networking platform. The post is published at the optimal time based on the algorithm and user trends of the selected platform.

[0684] Step 6:

[0685] Users view posts on social media and react to them. Engagements such as comments and "likes" indicate the user's level of engagement with the post and influence subsequent server processing.

[0686] Step 7:

[0687] The server collects performance data for each post and analyzes engagement metrics (e.g., follower growth rate, likes, shares). This analysis is then used to improve future post creation and marketing strategies.

[0688] (Example 1)

[0689] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0690] The problem this invention aims to solve is to automate everything from data collection to content generation, publication, and user engagement optimization in order to conduct effective marketing activities on social media. In particular, it is necessary to process large amounts of data in real time and use natural language processing to filter and normalize it, thereby enabling the generation of effective content based on lean information. Furthermore, it is necessary to review and optimize the generated content and provide automated responses to maximize user engagement.

[0691] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0692] In this invention, the server includes information gathering means, preprocessing means, generative model learning means, information generation means, information publication means, communication means, analysis means, means for reviewing and optimizing the generated content, and means for providing automated responses to maximize user engagement. This enables efficient and effective marketing activities on social media, real-time data processing, optimization of information through natural language processing, and interaction with users.

[0693] An "information gathering device" is a device that has the function of acquiring information from multiple sources via a communication network and collecting data in real time.

[0694] A "preprocessing device" is a device that filters out unnecessary information from acquired data and then normalizes the data using natural language processing.

[0695] A "generative model learning device" is a device that uses pre-processed data to train a generative AI model and acquire patterns to maximize user engagement.

[0696] An "information generation means" is a device that automatically generates new information using a pre-trained generative model and creates content for posting.

[0697] "Information disclosure means" refers to a device that publishes generated content on a designated online platform and posts it considering the optimal time and hashtags.

[0698] A "communication device" is a device used to send and receive data and instructions in real time between a server and a terminal.

[0699] An "analysis tool" is a device that analyzes data such as user responses and changes in follow-up activity to extract information that can be used to improve future marketing activities.

[0700] "Means for reviewing and optimizing generated content" refers to a system that builds an AI-driven feedback loop to automatically improve the quality of generated content.

[0701] "Means of providing automated responses to maximize user engagement" refers to a device that automatically provides appropriate responses to user comments, thereby deepening interaction with the user.

[0702] This invention is an advanced automated system for effectively conducting marketing activities on social media using an AI agent. This system consists of three main elements: a server, a terminal, and a user. The roles and specific usage methods of each are described below.

[0703] Server Role

[0704] The server plays a central role in everything from information gathering to processing, generation, and analysis. First, the server uses information gathering tools to acquire data from multiple sources via the communication network. This data includes past posts, current trends, and the activities of competitors. Next, the server uses Python's Pandas and natural language processing tools as preprocessing tools to filter and normalize the data and remove unnecessary information.

[0705] Subsequently, the system learns patterns in the collected data using generative model training methods such as TensorFlow or PyTorch. This training explores the optimal methods for increasing user engagement. This trained model is then used as an information generation tool, automatically generating content using a text generation model (e.g., GPT). Furthermore, the server reviews the generated content and improves its quality through an AI feedback loop for optimization.

[0706] Terminal role

[0707] The terminal is responsible for publishing generated content to the SNS platform based on instructions sent from the server. The terminal receives data in real time using the WebSocket protocol and posts content considering the optimal time and hashtags. This allows for efficient reach to the target user base.

[0708] User roles

[0709] Users view posts through their devices and can comment, like, and share them. This interaction is fed back to the server via communication methods and analyzed to inform future marketing activities.

[0710] Examples of specific cases and prompt statements

[0711] For example, in the promotion of a new product, a server analyzes the latest market trends and generates highly relevant content. This content is posted on social media in a format that suggests the product's unique features and usage scenarios. Users react to this content, and this data is analyzed and used to generate subsequent content.

[0712] An example of a prompt might be: "Generate a social media post about the next new product based on the latest market trends. Maximize follower engagement."

[0713] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0714] Step 1:

[0715] The server retrieves data from multiple sources using information gathering tools. The input consists of data provided by each source. This data includes social media posts, trend information, and competitor activities. APIs are used to collect information in real time during data acquisition. The output is a collection of the retrieved raw data.

[0716] Step 2:

[0717] The server cleans the raw data obtained using preprocessing methods. The input is the raw data obtained in step 1. Specifically, the Pandas library in Python is used to filter out spam data and noise and normalize the data. Natural language processing tools are also used to format the text data. The output is refined, clean data.

[0718] Step 3:

[0719] The server uses a generative model training method to train a model on clean data. The input is the purified data obtained in step 2. This training is performed using machine learning libraries such as TensorFlow and PyTorch, and explores patterns that enhance user engagement. The output of this process is a trained generative AI model.

[0720] Step 4:

[0721] The server utilizes information generation tools and generates new content using a trained generative model. The input consists of the trained model and prompt text obtained in step 3. Specifically, a text generation model (e.g., GPT) automatically generates content for posting based on the instructions in the prompt text. This output becomes the text data to be posted.

[0722] Step 5:

[0723] The device publishes the content generated via the information disclosure mechanism on the SNS platform. The input is the generated content obtained in step 4. The device sends the content using the WebSocket protocol, taking into account the appropriate time of day and relevant hashtags. The output is the post published on SNS.

[0724] Step 6:

[0725] Users view published content and leave comments and reactions. Input is posts on social media. User interactions are fed back to the server, which is used for future data collection and content generation. Output is user response data.

[0726] Step 7:

[0727] The server uses analytical tools to analyze user response data and gain insights to improve future marketing strategies. The input is the user response data obtained in step 6. The analysis evaluates the effectiveness of the content using metrics such as follower growth / decrease and engagement rate. The output is a proposal for an improved marketing strategy.

[0728] (Application Example 1)

[0729] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0730] In today's advertising industry, there is a demand for personalized ads that are displayed at the right time based on users' interests and preferences. However, traditional advertising systems struggle to generate and display ads that respond to user behavior in real time, which limits the effectiveness of marketing activities.

[0731] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0732] In this invention, the server includes an information gathering device, a data processing device, and a generative model learning device. This makes it possible to acquire user behavior data in real time and generate and display advertising content optimized for the user based on that data.

[0733] An "information gathering device" is a device used to acquire data from individual information sources via an information network.

[0734] A "data processing device" is a device that analyzes collected data and transforms it into an optimized form based on user behavior and interests.

[0735] A "generative model learning device" is a device that uses processed data to build and train a learning model for generating advertising content based on user interests.

[0736] An "information generation device" is a device that has the function of actually generating advertisement copy and content using a trained model.

[0737] An "information transmission device" is a device that delivers generated advertising content to users through a user interface.

[0738] A "mutual communication device" is a device that transmits user responses and feedback to a server and uses them as data for generating the next content.

[0739] A "data analysis device" is a device that analyzes user response data and provides information for optimizing marketing strategies.

[0740] A "user interface device" is a device that allows users to directly interact with advertising content visually.

[0741] A "personalized display device" is a device that displays individually optimized advertisements to users according to their needs and interests.

[0742] To realize this invention, the system has the following configuration: The server operates with an information gathering device, a data processing device, and a generative model learning device. The information gathering device is responsible for collecting user data from social networking services (SNS) and other information sources via an information network. The data processing device cleanses the collected data and organizes it appropriately based on the user's interests and behavior. This process includes filtering spam data and removing unnecessary information.

[0743] Next, the generative model learning system uses the cleansed data and machine learning libraries such as TensorFlow to train a model for generating advertising content optimized for user interests. This enables the generation of new patterns based on past successful advertisements.

[0744] The terminal has a user interface device and is responsible for displaying generated advertising content to the user. When a user views an advertisement using the terminal, their response is sent to the server via a communication device. This response is analyzed in real time by the server's data analysis device and used to generate content for the next time.

[0745] For example, if a user frequently views sports-related posts, the generative model will use that data to create advertisements for the latest sports equipment. A possible prompt might be, "Based on the following information, create an ad copy that is likely to interest the user: The user's profile shows a lot of sports activity and they have liked many running-related posts." In this way, more appropriate and effective marketing activities become possible.

[0746] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0747] Step 1:

[0748] The server uses information gathering devices to collect user-related data from the information network. This data includes information about the user's activity history and interests on social networking services (SNS). It uses data obtained through SNS APIs as input and outputs it in a standardized format.

[0749] Step 2:

[0750] The server uses a data processing unit to cleanse the collected data. This process filters out spam and irrelevant information and extracts characteristic data based on user behavior. The input is the collected raw data, and the output is clean, analyzable data.

[0751] Step 3:

[0752] The server uses a generative model learning system to train a generative AI model using cleansed data and tools such as TensorFlow. In this step, patterns are identified to optimize user engagement. The input is the analyzed data obtained in step 2, and the output is the trained AI model.

[0753] Step 4:

[0754] Using the generated model, the server generates personalized advertising content tailored to the user's interests through an information generator. Here, ad copy is created using prompts. The input is a trained AI model and prompts, and the output is the specific ad content.

[0755] Step 5:

[0756] The terminal uses a user interface device to display generated advertising content to the user. This display is done via a smartphone or smart glasses. The input is advertising content received from the server, and the output is the advertisement visually presented to the user.

[0757] Step 6:

[0758] When a user interacts with an advertisement via their device, their response is sent to the server via a communication device. This response includes the number of clicks on the advertisement and comments. The input is user behavior data, and the output is feedback data for future content generation.

[0759] Step 7:

[0760] The server uses data analysis equipment to analyze user feedback and use it to improve advertising strategies. These analysis results are then incorporated into the creation of new advertising content. The input is feedback data, and the output is information for improving advertising strategies.

[0761] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0762] This invention incorporates an emotion engine that recognizes user emotions into a system that utilizes an AI agent to conduct effective marketing activities on social media. This allows for real-time analysis of the user's emotional state and optimization of marketing activities based on this analysis. The specific functions and usage examples for each element—server, terminal, and user—are described below.

[0763] Server Role

[0764] The server acts as the central hub of the system, handling data collection, processing, generation, and analysis. It also utilizes an emotion engine to analyze users' emotional states. Initially, the server collects posted data from social media platforms using various information gathering methods. By focusing on collecting data from particularly popular posts, comments, and reactions, it obtains a wealth of data regarding users' emotions.

[0765] Next, the data is cleaned using a preprocessing mechanism to organize it into a state suitable for training the generative model. Unnecessary data is removed during this process. The generative model training mechanism then performs model training incorporating sentiment data based on the preprocessed data. This enables content generation that takes into account changes in the user's emotions.

[0766] The emotional data acquired by the emotion engine helps generate appropriate content for users through information generation methods. For example, it can create encouraging messages for users who show positive reactions and supportive posts for users who show negative reactions.

[0767] Terminal role

[0768] The device publishes content generated from the server to the social networking platform at the appropriate time. This includes scheduling and adjusting posts based on sentiment analysis results. By posting during times when users are most active, engagement can be maximized.

[0769] User roles

[0770] Users interact with content provided through social media platforms. User comments and reactions are analyzed in real time using an emotion engine, and feedback is immediately provided by agents. Understanding each user's individual emotional state enables personalized responses.

[0771] Specific example

[0772] For example, during a campaign announcing a product launch, the server learns typical user response patterns from relevant historical data and continuously monitors users' real-time reactions using an emotion engine. In this case, it automatically generates posts offering coupons to users with positive emotions and provides special support guidance to users with negative emotions, thereby improving the overall user experience.

[0773] Thus, the present invention is a system that can realize more personalized marketing activities that take user emotions into consideration.

[0774] The following describes the processing flow.

[0775] Step 1:

[0776] The server collects post data related to predefined keywords and hashtags through the SNS platform's API. It prioritizes collecting posts that have garnered particularly high attention and stores this data in a database.

[0777] Step 2:

[0778] The server preprocesses the collected data. This process filters out unnecessary information and spam, extracting important data. It also cleans the text necessary for sentiment analysis and reduces noise.

[0779] Step 3:

[0780] The server trains a generative model using pre-processed data, while simultaneously using a sentiment engine to analyze sentiment data for each post and comment. The sentiment engine distinguishes between positive, negative, and neutral sentiments, tags them, and stores them.

[0781] Step 4:

[0782] Based on generative model training, the server automatically generates new posts that reflect sentiment data. Here, messages are adjusted according to the user's emotions, and content is constructed with an appropriate tone.

[0783] Step 5:

[0784] The system optimizes the timing of when the device receives content generated from the server and publishes it to the social networking platform. During this process, the posting schedule is adjusted based on the user's activity time and past posting patterns.

[0785] Step 6:

[0786] Users comment on and react to posts on social media, and their emotional states are recorded within the system. User feedback is reflected in real time and analyzed by the system's emotion engine.

[0787] Step 7:

[0788] The server analyzes user reaction data to posts and generates engagement metrics. This data is used to adjust future content creation and marketing strategies, optimizing the entire system.

[0789] (Example 2)

[0790] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0791] In today's digital society, effective marketing activities that take user emotions into account require the rapid and accurate collection of data from diverse sources and the selection of useful information from that data. However, traditional methods are inefficient in data collection and analysis, making it difficult to grasp user emotions in real time and reflect them in marketing strategies.

[0792] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0793] In this invention, the server includes an information acquisition means, a numerical processing means, and an emotion analysis means. This makes it possible to quickly collect necessary information from a large amount of data on an SNS platform, effectively remove unnecessary data, and then accurately analyze the user's emotions. This enables personalized marketing activities that respond to changes in the user's emotions.

[0794] "Information acquisition means" refers to a mechanism for obtaining necessary data from multiple information sources using wide-area communication.

[0795] "Numerical processing means" refers to functions that remove noise and unnecessary data from acquired data and prepare it in a format suitable for analysis.

[0796] "Computational model learning methods" refer to the process of training machine learning models using pre-processed data to obtain insights that are useful for future predictions and generation.

[0797] "Information creation means" refers to methods for generating new information or content based on a learned model.

[0798] "Information distribution means" refers to the function of publishing generated information to the appropriate platform at the optimal time.

[0799] "Information transmission means" refers to communication circuits and protocols used to exchange data within and outside a system.

[0800] "Evaluation methods" refer to methods for analyzing the performance of generated information and the overall system to identify areas for improvement.

[0801] "Emotional analysis tools" are functions that extract the user's emotional state from data and classify and analyze it.

[0802] This invention is a system based on advanced information technology that analyzes user emotions on a social networking service (SNS) platform and enables personalized marketing based on that analysis. The system is composed of an information acquisition means, a numerical processing means, an emotion analysis means, a computational model learning means, an information creation means, an information distribution means, and an information transmission means.

[0803] The server first acquires a large amount of user data from SNS platforms and other sources via wide-area communication using information acquisition methods. Specifically, it collects data primarily from posts, comments, and reactions that attract particularly high attention through API access. Based on this data, the server performs preprocessing using numerical processing methods to remove noise and unnecessary data, preparing it for analysis. Open-source libraries (e.g., NLTK and spaCy) can be used as natural language processing tools.

[0804] Next, the server analyzes the sentiment of the pre-processed data using sentiment analysis tools. Existing natural language processing APIs (e.g., Google Cloud Natural Language API) are used for sentiment analysis, classifying the user's emotional state with tags such as positive, negative, and neutral. Based on these analysis results, a generative AI model is trained using computational model learning tools. TensorFlow and PyTorch are expected to be used as machine learning frameworks.

[0805] The server uses the acquired emotion data to generate content that responds to the user's emotions using an information creation mechanism. For example, it is possible to input a prompt such as, "Generate an encouraging message based on the user's positive response data," and the AI ​​can dynamically generate a message.

[0806] The generated content is published to social media platforms via information distribution methods at the optimal time determined by the device. The device analyzes user activity patterns and adjusts posts to coincide with the time when engagement is maximized. This can enhance the overall effectiveness of marketing activities.

[0807] Users react to and comment on the delivered content. This feedback is sent back to the server via communication channels and used for sentiment analysis and content generation in the next round. The learning process of the generative AI model is continuously improved by incorporating user feedback, enabling the delivery of more effective marketing messages.

[0808] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0809] Step 1:

[0810] The server uses information acquisition methods to obtain large amounts of user data from SNS platforms via wide-area communication. The input is raw posted data via API, and the output is a collection of raw data organized in chronological order. At this stage, priority is given to collecting posts with a large number of "likes" and "shares."

[0811] Step 2:

[0812] The server performs preprocessing on the acquired raw data using numerical processing tools. The input is the raw data obtained in step 1, and the output is the cleaned-up data. This preprocessing includes removing noise data and spam posts, and tokenizing comments and posts. The text is morphologically analyzed using a natural language processing library and formatted into a format suitable for analysis.

[0813] Step 3:

[0814] The server uses sentiment analysis tools to analyze pre-processed data and classify the user's emotional state. The input is cleaned-up data, and the output is data tagged with sentiment (positive, negative, neutral). A dedicated API is used for sentiment analysis, and accuracy is improved by scoring the intensity of each emotion.

[0815] Step 4:

[0816] The server uses computational model learning methods to train a generative AI model based on emotion-tagged data. The input is emotion-tagged data, and the output is a model that can predict changes in the user's emotions. A machine learning framework is used to extract features from the data and create an optimized model.

[0817] Step 5:

[0818] The server generates content using a trained model and prompt text via an information creation mechanism. The input is the trained model and the prompt text "Generate an encouraging message based on the user's positive response data," and the output is a personalized message. The AI ​​suggests the most appropriate wording based on the user's past response patterns.

[0819] Step 6:

[0820] The device publishes generated content to social media platforms at the optimal time via an information distribution method. The input is the generated content, and the output is a successful posting message to social media. The device analyzes past user activity times and delivers content at the time when engagement is highest.

[0821] Step 7:

[0822] Users provide feedback by reacting to and commenting on published content. The input is the content published on social media, and the output is new comments and reactions from users. This feedback is sent back to the server and used for sentiment analysis and content generation in the future.

[0823] (Application Example 2)

[0824] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0825] In marketing activities on social media, there is a challenge in that personalized ad delivery that takes into account the individual emotions of users is not being adequately implemented. As a result, ads that do not match users' interests or emotions are often delivered, which can lead to decreased ad effectiveness and a decline in the quality of the user experience.

[0826] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0827] In this invention, the server includes information gathering means, sentiment analysis means, and advertisement generation means. This makes it possible to analyze the user's emotions based on data collected from the SNS platform and deliver advertisements that are appropriate to the user's emotions in real time based on the analysis results.

[0828] "Information gathering means" refers to a system that has the function of acquiring information from multiple sources via a communication network.

[0829] A "preprocessing means" is a device that has the function of filtering out unnecessary information from the acquired information and using the refined information.

[0830] A "generative model learning tool" is a device that has the function of enabling an AI model to learn based on pre-processed data.

[0831] An "information generation means" is a device that has the function of generating appropriate content based on user sentiment data.

[0832] "Information disclosure means" refers to a system that has the functionality to publish generated content on a social networking service (SNS) platform.

[0833] "Communication means" refers to anything that has the function of sending and receiving data between each element within a system.

[0834] "Analysis tools" are those that analyze collected data and have the function of revealing users' emotions and behavioral patterns.

[0835] An "emotion analysis tool" is a device that has the function of analyzing the user's emotions from the collected data.

[0836] An "advertising generation method" is a device that has the function of generating advertisements appropriate to the user based on the results of sentiment analysis.

[0837] The server uses information gathering means to collect posted data from SNS platforms. In this process, it obtains the latest user posts and reaction data from Twitter and Facebook APIs via a communication network. This collected data is then filtered and refined by preprocessing means. Based on this refined data, a generative model learning means performs user sentiment analysis using an AI model (e.g., BERT or GPT).

[0838] The results of the sentiment analysis are used by the ad generation system to generate ads that are appropriate to the user's emotions. These ads are then prepared for posting to social media platforms via the information generation system and published at the appropriate time by the information publication system.

[0839] The device publishes this advertisement on social media based on instructions from the server and displays it at the appropriate time to increase user engagement. Users react to the published advertisement, which collects even more sentiment data that is used as training data for the entire system.

[0840] For example, if a user posts "Today is the best day!" on social media, the server analyzes this as a positive emotion and generates a restaurant coupon for a new product using an ad generation tool. Ultimately, this ad will appear in the user's feed.

[0841] Example prompt: "For posts that show positive user emotions, display ads that enhance enjoyment; for posts that show negative emotions, display ads that soothe the soul."

[0842] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0843] Step 1:

[0844] The server uses information gathering tools to obtain posted data and reaction data from the SNS platform's API via the communication network. In this process, messages and comments generated by users on SNS are collected as input, and the data is output in a standard data format for use in subsequent data processing.

[0845] Step 2:

[0846] The server uses preprocessing to filter out unnecessary data from the data collected in step 1 and extract the necessary information. In this process, since the input data may contain noise or duplicate data, these are removed, and clean data in a format suitable for sentiment analysis is output.

[0847] Step 3:

[0848] The server passes the refined data to a generative model training tool, which then uses an AI model (e.g., BERT or GPT) to analyze the user's emotions. This data calculation outputs a user's emotion score (positive, negative, neutral) based on the input clean data.

[0849] Step 4:

[0850] The server receives the output from the sentiment analysis system and uses the ad generation system to create ads that match the user's emotions. Here, the user's sentiment score is used as input, and highly relevant ad campaigns and promotional information are output.

[0851] Step 5:

[0852] The terminal uses information generation means to prepare to post the advertisement generated in step 4 to the SNS platform. In this preparation stage, the input advertisement data is converted into an appropriate format and output with scheduling information added.

[0853] Step 6:

[0854] The device publishes advertisements generated at the appropriate time on social media through an information disclosure mechanism. Here, posts prepared using the social media API are used as input data, timely publication processing is performed, and the result of the advertisement appearing in the user's feed is output.

[0855] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0856] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0857] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0858] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0859] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0860] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0861] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0862] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0863] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0864] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0865] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0866] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0867] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0868] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0869] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0870] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0871] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0872] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0873] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0874] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0875] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0876] The following is further disclosed regarding the embodiments described above.

[0877] (Claim 1)

[0878] Information gathering methods,

[0879] Pre-treatment means,

[0880] Generative model learning method,

[0881] Information generation means and

[0882] Information disclosure methods,

[0883] Communication methods,

[0884] Analytical methods,

[0885] A system that includes this.

[0886] (Claim 2)

[0887] The system according to claim 1, wherein the information gathering means has the function of acquiring information from multiple information sources via a communication network.

[0888] (Claim 3)

[0889] The system according to claim 1, wherein the preprocessing means has the function of filtering out unnecessary information from the acquired information and using the purified information.

[0890] "Example 1"

[0891] (Claim 1)

[0892] Information gathering methods,

[0893] Pre-treatment means,

[0894] Generative model learning method,

[0895] Information generation means and

[0896] Information disclosure methods,

[0897] Communication methods,

[0898] Analytical methods,

[0899] A means of reviewing and optimizing the generated content,

[0900] A means of providing automated responses to maximize user engagement,

[0901] A system that includes this.

[0902] (Claim 2)

[0903] The system according to claim 1, wherein the information gathering means has the function of acquiring information from multiple information sources via a communication network and also performs real-time data collection.

[0904] (Claim 3)

[0905] The system according to claim 1, wherein the preprocessing means has the function of filtering out unnecessary information from acquired information and normalizing the data using natural language processing.

[0906] "Application Example 1"

[0907] (Claim 1)

[0908] Information gathering device,

[0909] Data processing device,

[0910] Generative model learning device,

[0911] Information generation device and

[0912] Information transmission device,

[0913] Intercommunication device,

[0914] Data analysis device,

[0915] User interface device and

[0916] Personalized display devices,

[0917] A system that includes this.

[0918] (Claim 2)

[0919] The system according to claim 1, wherein the information gathering device has the function of acquiring information from multiple information sources via an information network and processing information based on user behavior in real time using the data processing device.

[0920] (Claim 3)

[0921] The system according to claim 1, wherein the generative model learning device has a function of generating advertising content to be displayed on the personalized display device using information refined by the data processing device.

[0922] "Example 2 of combining an emotion engine"

[0923] (Claim 1)

[0924] Means of obtaining information,

[0925] Numerical processing means,

[0926] Computational model learning methods,

[0927] Information creation methods,

[0928] Information distribution methods,

[0929] Means of information transmission,

[0930] Evaluation methods,

[0931] Emotion analysis methods,

[0932] A system that includes this.

[0933] (Claim 2)

[0934] The system according to claim 1, wherein the information acquisition means has the function of acquiring information from a large number of information sources via wide-area communication.

[0935] (Claim 3)

[0936] The system according to claim 1, wherein the numerical processing means has the function of removing unnecessary data from the acquired information and utilizing the selected information.

[0937] "Application example 2 of combining emotional engines"

[0938] (Claim 1)

[0939] Information gathering methods,

[0940] Pre-treatment means,

[0941] Generative model learning method,

[0942] Information generation means and

[0943] Information disclosure methods,

[0944] Communication methods,

[0945] Analytical methods,

[0946] Emotion analysis methods,

[0947] means of generating advertisements,

[0948] A system that includes this.

[0949] (Claim 2)

[0950] The system according to claim 1, wherein the information gathering means has the function of acquiring information from multiple information sources via a communication network.

[0951] (Claim 3)

[0952] The system according to claim 1, wherein the preprocessing means has the function of filtering out unnecessary information from the acquired information and using the purified information. [Explanation of Symbols]

[0953] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

[Claim 1] Information gathering means that acquire information from multiple sources via a communication network, A preprocessing means for filtering out unnecessary information from the acquired information, A generative model learning method that uses filtered data to train a generative model based on past successes, An information generation means that generates new, attention-grabbing content using a pre-trained generative model, Information disclosure means for publishing generated content on an appropriate online platform, A means of communication that facilitates interaction and communication with users and promotes their engagement, Analytical methods for analyzing follower behavior and engagement after posting, A system that includes this.

Images