Methods, systems, and non-temporary computer-readable media for customizing community user robots.

Abstract

This provides methods, systems, and non-temporary computer-readable media for customizing community user robots. [Solution] The system provides a cloud server, and community users can upload data via the user interface provided by the cloud server to set the attributes of the community robot (including prompts for various attributes). Subsequently, a training dataset is formed after processing the data by a language model, and a machine learning method operating within that dataset is trained to learn the attributes of community users from the training dataset. Based on the prompts, the robot is trained to become a community robot dedicated to the community user, deployed to the community, and allowed to operate under the community user's personal account within the community.

Description

Technical Field

[0001] The present invention relates to creating a robot by learning user characteristics, and particularly to a method, system, and non - transient computer - readable medium for customizing a community user robot that creates an AI robot with user characteristics by learning data provided by a user.

Background Art

[0002] In artificial intelligence (AI), which is rapidly developing in various fields at present, the one that has developed greatly is generative artificial intelligence that realizes natural language processing by utilizing a large language model (LLM). One of its applications is a natural - language chatbot that can communicate with users in natural language.

[0003] Social media is a network service in which multiple people can share content, exchange messages, build friendships, and establish groups. Users joining the community can express opinions and interact with various messages, and provide a place for interaction with various forms of content such as text, images, music, and videos. [[ID=​​​​​Traditional social media primarily provides services that allow community users to engage in activities within their communities. AI robots operating on well-known social media platforms mainly provide services that answer questions via chat (for example, customer service for companies), and by learning the characteristics of community users, they can conduct accurate marketing. [Overview of the project] [Problems that the invention aims to solve]

[0006] The present invention provides a method, system, and non-temporary computer-readable media for customizing community user robots, and unlike current applications using community robots, the method provided enables community users to train community robots belonging to them to perform activities within their community. [Means for solving the problem]

[0007] According to the embodiment, the method for customizing the provided community user robot is executed on a cloud server, which operates the community. The main steps of the method include receiving data provided by community users, performing preprocessing on the data including processing the data with a language model to form a training dataset, then training using the training dataset with a machine learning method to create a community robot dedicated to the community user, thereby deploying the community robot to the community, creating a link with the community user's personal account, and allowing the community robot to operate under the community user's personal account within the community.

[0008] Furthermore, community users can upload data via a user interface launched within the application program, including documents, graphics, video content, and website or channel links. In addition, the user interface is used not only to allow community users to upload data, but also to enable them to style the community robot, configure profiles, provide personal data, set up community activities, and perform back-office management.

[0009] Furthermore, the aforementioned data is personal data relating to the community users, which is processed by a language model to obtain text for training a machine learning method, forming a user feature dataset that describes the community users, and a community robot dedicated to the community users is created so that the machine learning method can be used to learn the personal characteristics of the community users.

[0010] Furthermore, the data may include video data or other information of community users, and combined with the personal characteristics of community users obtained through learning, the generative artificial intelligence can generate simulated images of community robots.

[0011] Furthermore, the community robot, which is exclusively for the community user, can automatically operate within the community and process the community user's community activities within that community. [Brief explanation of the drawing]

[0012] [Figure 1] A schematic diagram illustrating one scenario in which a customized community user robot is utilized. [Figure 2] A schematic diagram showing an example of a system configuration for customizing community user robots. [Figure 3]A flowchart illustrating an example of how to customize a community user robot. [Figure 4] A diagram showing a first embodiment of a user interface for customizing community user robots provided to community users. [Figure 5] A diagram showing a second embodiment of a user interface that allows community users to customize community user robots. [Figure 6] This figure shows a third embodiment of a user interface that allows community users to customize community user robots. [Figure 7] This figure shows a fourth embodiment of a user interface that allows community users to customize community user robots. [Figure 8] A diagram showing a first embodiment of a user interface utilizing a community user robot. [Figure 9] A diagram showing a second embodiment of a user interface utilizing a community user robot. [Figure 10] A diagram showing a third embodiment of a user interface utilizing a community user robot. [Modes for carrying out the invention]

[0013] The present invention provides a method for customizing a community user robot, a system for customizing a community user robot, and non-temporary computer-readable media, the method of which is executed on a cloud server, where community users can customize a community robot dedicated to them personally by operating a user interface over a network, and by deploying the trained community robot to the community, the community robot can perform community activities based on training datasets provided by community users.

[0014] According to an embodiment of a method for customizing community user robots, users (community users) can order a custom-made AI robot just for them. Users on terminals can customize their own AI robots via a server, and in particular, the AI ​​robots can reflect the individual user's expertise and knowledge. For example, an influencer can create an AI robot that can interact with internet users on their behalf. The machine learning algorithm (natural language processing module) executed by the system learns the characteristics of the data provided by the user and creates a customized AI robot for that user.

[0015] Figure 1 is a schematic diagram illustrating one scenario in which a customized community user robot is utilized.

[0016] The figure shows a cloud server 20 that manages the community while executing methods for customizing community user robots. The cloud server 20 is implemented by a computer system, and processing circuits within the computer system execute machine learning algorithms related to the customization of community user robots. It can train community robot models based on training datasets containing the attributes of individual community users. These AI robots may be generated by Generative Artificial Intelligence based on various knowledge, conversations, actions, and facial expressions related to the attributes of individuals in the training dataset, or they may be pseudo-robots with a human shape or specific images (people, anime, animals, objects). As shown in the figure, an AI robot 101 with a pseudo-image is running on a user device 100 within the community. Each AI robot 101 may be a community robot trained based on data provided by a particular community user within the community. The AI ​​robot 101 can interact with users in the community in the form of text, voice, images, and video (including the ability to converse with user 10 in natural language). According to the embodiment, during the process of natural language dialogue, the smart model operating behind the AI ​​robot 101 can learn the pitch of the user's voice frequency and speaking speed while engaging in voice dialogue with the user, enabling the AI ​​robot 101 to engage in dialogue with appropriate voice frequencies and speaking speed, and to reflect the user's current emotions in a timely manner.

[0017] According to the embodiment, if the smart model analyzes changes in tone, rhythm, and voice intensity in the voice data and determines that user 10 is currently sad, the AI ​​robot 101 can adjust the voice frequency and speaking speed of the conversation in response to this sad emotion, thereby reflecting user 10's current sad emotion in a timely manner.

[0018] The figure schematically shows how user 10 operates a community application program or an application program that can access social media on user device 100. For example, after opening a user interface via the community application program and connecting to cloud server 20, the user selects social media content from cloud server 20 and loads a plurality of AI robots 101 (which may include community robots trained by the user 10 himself / herself in a way of customizing the community user robots executed by cloud server 20) displayed on user device 100 as shown in the figure. Here, relevant model data and video signals of each AI robot 101 are included, enabling user 10 to experience something like being in a video call with two people in the real world. The illustrated embodiment in the figure shows starting a user interface on the display of user device 100, using the camera lens 105 on user device 100 to capture user 10 himself / herself, presenting a user image 103 as shown in the figure, and the background video being a plurality of AI robots 101 deployed on social media.

[0019] When engaging in community activities, the user device 100 connects to the cloud server 20 and opens the community application program on the user device 100. According to one embodiment, the community robot can independently publish content on social media, view and like Moments content, share it, and comment on it. In addition, it can activate audio (microphone) and video (camera) functions, allowing the user 10 to interact with one of the AI ​​robots 101. The robot can collect sound and capture video in real time, upload the audio and video data to the cloud server 20 in real time, and use language and video models running on the cloud server 20 to capture audio characteristics (or video characteristics, and possibly current environmental information). Utilizing smart models, the robot can determine the user 10's current emotions, generate voice dialogue, and deliver the voice through the AI ​​robot 101, or provide content feedback in any format. Here, the basis for the smart model's ability to determine emotion includes analyzing the tone, rhythm, and intensity of each syllable in the speech data, as well as the meaning of the words.

[0020] The system for customizing community user robots provided by the present invention utilizes technologies such as natural language models, machine learning methods, and generative artificial intelligence via a computer system to create AI robots that can operate on social media. In particular, community robots created for community users possess the attributes of the community users themselves, form virtual users that operate on social media, can engage in community activities with real users participating in social media, and can even provide generated community content.

[0021] For the system created by the cloud server 20 shown in FIG. 1, reference can be made to the schematic diagram of an embodiment related to the configuration of a system for customizing the community user robot shown in FIG. 2. The cloud server 20 can create and operate a community, and the community robot created by the method can also operate in a community operated by another external platform. According to the embodiment shown in FIG. 2, the cloud server 20 enables the user of the terminal to join the community user through the network 200, and by the method of customizing the community user robot executed therein, the community user generates a community robot exclusive to the individual community user.

[0022] The examples shown in the figure illustrate the provision of services to the user devices 201, 202, and 203 of the terminal users via the network 200. Here, a community robot in anthropomorphic form is shown on the screen of the user device 201. On the screen of the user device 202, a plurality of community robots operating in the community are shown, and community activities such as group chat, meeting, and live broadcast can be carried out. The community robot shown on the user device 203 can conduct a series of conversations with other community users.

[0023] According to the example, through the collaboration of software and hardware on the cloud server 20, a community robot model is operated, allowing each community user to have a dedicated community robot that autonomously manages text, graphic, or video channels on their behalf, thereby building community relationships and providing community users with a means to build their own followers and also become followers of others. Furthermore, it becomes possible to create groups within social media such as fan pages, community groups, and blogs, enabling interaction with other real or virtual users on social media.

[0024] The system for customizing community user robots, implemented by the cloud server 20, implements a certain platform that allows community users to customize their own personal community robots. According to one embodiment, the cloud server 20 has a database 25 built-in or externally attached, and the cloud server 20 accesses the database 25 via its database module 215. The database 25 can contain user data 251 for storing community user data and artificial intelligence pseudo-human models 253. The artificial intelligence pseudo-human models 253 may include models of various community robots created by community users using the method described above, and pseudo-human video data formed using generative artificial intelligence technology.

[0025] According to the schematic diagram of an embodiment of the system configuration shown in Figure 2, the cloud server 20 includes a processing circuit 21, memory 23, and various functional modules realized through the collaboration of software and hardware (processing circuit 21, memory 23, and related circuits), and also includes a natural language processing module 211. Here, a language model 211a and a speech model 211b, an instruction processing module 212, a user interface module 213, and a machine learning module 214 can be run, and here, a generative artificial intelligence 214a and an AI robot model 214b, a database module 215, and a community module 216 can be run.

[0026] In this embodiment, the cloud server 20 operates the community via the community module 216, processing data generated by the activities of community users and community robots within the community. This allows each community user and community robot to have their own subscribers and followers within the community, and to like, dislike, reply to, and share automatically generated content. Furthermore, the natural language processing module 211 is used to process messages from various interactions generated within the community (e.g., dialogue content between community users). Natural language generation (NLG) is implemented by the language model 211a, enabling the generation of natural, fluent, and situationally appropriate text. Additionally, by processing audio data with the voice model 211b (for example, simulating the voice of a community user and communication with other community users), community robots can generate and send messages. The cloud server 20 facilitates text, audio, and video interactions via a graphical user interface (GUI), allowing community users to create their own personal community robots and displaying community content based on user input.

[0027] Furthermore, within the community operated by the cloud server 20, the language model 211a and speech model 211b, which operate using the natural language processing module 211, can process community data generated within the community, acquire features from that community data, and process the semantics of comments or dialogues from community users or community robots within the community, enabling community users to acquire content that suits their preferences.

[0028] According to the embodiment, the community robot created by the method of customizing the community user robot is capable of operating within the community and can also build a specific professional background through training. Preferably, the community user provides knowledge data on their field of expertise or hobbies and preferences, and the AI ​​robot model 214b in the machine learning module 214 is trained to become a community robot equipped with the community user's knowledge. Separately, the generating artificial intelligence 214a operates Retrieval Augmented Generation (RAG) technology to realize a community robot that is provided for a specific field based on a Large-Scale Language Model (LLM).

[0029] Furthermore, the generating artificial intelligence 214a is trained to match the appearance of the community user or the community robot they require, based on text, image, and video data provided by the community user, and generates a simulated video of a community robot operating within the community. According to one embodiment, the generating smart model on which the system for customizing community user robots operates can employ (but is not limited to) Google's VLOGGER AI technology, which can generate videos that convey content by adding text and audio based on photos uploaded by the user. According to Google, VLOGGER AI technology is a multi-modal artificial intelligence with audio and dynamic video that can process complex facial expressions and limb movements to simulate videos of a person (or animal, anime, etc.) that can speak natural language.

[0030] In this way, the generating artificial intelligence 214a can determine the movable muscles (e.g., facial muscles), the five senses, and the parts that can express emotions in the uploaded video, and automatically generate a simulated video of a community robot operating in the community. Combined with the attributes of the community users that the community robot has trained on based on data provided by community users, it can interact on behalf of community users and generate community content such as video content.

[0031] The instruction processing module 212 in the cloud server 20 is used to process data generated by community users' activities within the community, processing requests generated by each community user, analyzing each community user's needs, and promptly responding to the community robot. For example, a community user may be an influencer, a company's customer service, or an organization managing a channel, and its dedicated community robot (e.g., a customer service robot) can provide appropriate services in real time based on the needs generated by the community user through an interactive interface.

[0032] The user interface module 213 in the cloud server 20 is used to generate an interface for interaction between terminal users, community users, and community robots, and is connected to the user interface generated by a specific application program running on the user device. In the community, the cloud server 20 communicates with user devices 201, 202, and 203 via the user interface module 213 (including the transmission of various messages (e.g., video content, text content, and video files, etc.) including necessary codecs, compression, and decompression programs), and also processes the conversational content between one or more community users and community robots displayed through the user interface, which is used to generate various functions and graphics in the user interface.

[0033] According to the embodiment, user devices 201, 202, and 203 are like a specific type of computer device, where a processor performs various functions of the computer system, and other storage functions are performed while non-temporary computer-readable media is implemented by memory. The non-temporary computer-readable media stores a program set, and the processors of user devices 201, 202, and 203 execute the program set and then launch an application program (for example, an application program that manages a community). The application program executed on the user device connects to the cloud server 20 and causes community users to create their own community robots.

[0034] For information on how to customize the community user robots that run using the above system configuration, please refer to the flowchart of the embodiment shown in Figure 3.

[0035] In a method for customizing community user robots running on a cloud server, the system provides a user interface that allows community users to order custom-made community robots. Referring to the examples of user interfaces that allow community users to customize community user robots shown in Figures 4 to 7, the system is configured to receive data provided by community users (step S301).

[0036] According to an embodiment relating to data provided by community users, community users can receive prompts for setting the style, profile, and community activities of the community robot via the user interface 40. One embodiment is shown in Figure 4, in which a community user can provide data via the user interface 40 used to customize the community robot shown in Figure 4, and the user interface 40 provides several option settings 401 (for example, shown in the figure in the order of "Style", "Bio", "Brain", "Post", and "Access"). Taking "Style" as an example, the community user can input prompts 403 regarding the description of the community robot's appearance style into the field, provide an image of themselves or an image of the desired appearance, display the user's appearance 405 on the user interface 40, and separately provide a prompt 407 to generate the appearance.

[0037] Here, "Profile / Bio" is used to set the community robot's name, role, and self-introduction (which may also include other professional details). The "Post" function allows users to set the community robot's ability to post (for example, whether to have the community robot make daily updates, whether to automatically reply to messages from online users, etc.). The "Access" function is used to set the role the community robot will perform (for example, whether it belongs to the Human Resources (HR) department, the Digital Marketing department, the Retail Development department, etc.).

[0038] For example, community users can upload photos using the "Style" function and describe people or objects in the photos to train a machine learning model on the features of the photos. For instance, the model can acquire the five senses of a person in a photo (e.g., eyes, nose, and mouth, which are organs that can express emotions), and then use pixels to replace the image pixels of these organs, and further control their movement (e.g., change the image of the mouth over time when speaking, make the eyes blink, and make other parts change naturally) to achieve the goal of a simulated person. The prompt 403 for the community robot's "Style" could be "You can customize the appearance of the community robot based on the video data uploaded here, and select a voice that matches your image while adjusting the personality based on the uploaded text and image data," and the prompt 407 for generating the appearance could be "I am a woman with long, wavy dark brown hair, thick eyebrows, deep-set eyes with light brown irises, and a warm smile." In this way, community robots can be generated by generative artificial intelligence technology executed by a cloud server, after community users complete prompts, provide relevant data, and then adjust the functionality after generation.

[0039] Furthermore, in the method of customizing the community user robot, community users can upload data through a user interface launched in the application program. The types of data include documents, graphics, video content, and website or channel links. As shown in the embodiment in Figure 5, the function of the Brain 50 in the user interface 40 is shown here. In this way, community users can provide data about their professional background, hobbies, and preferences (for example, they can provide website data describing their own professional data via web links 501, upload data related to them via document uploads 503 (for example, their professional papers, writings, and publications), and provide social media managed by them via personal community channels 505). In this way, the generative artificial intelligence can learn the knowledge of the community user, and the correspondingly generated community robot can be endowed with the community user's knowledge to generate content that matches the community user's attributes.

[0040] Furthermore, the system allows community users to configure the style of the community robot, set a profile, provide personal data, configure community activities, and perform back-office management through a user interface. An example can be seen in Figure 6, where community users can set community-related attributes through the user interface 60 shown in Figure 6. These include Availability settings 601 used to set the community robot to be private / internal use or public, Voice settings 603 used to set voice attributes related to the community user themselves (e.g., male or female, cheerful, soothing, charming, etc.), and Personality settings 605 which allow the system to set the community user's personality (e.g., passionate, sweet, lively, charming, etc.). In addition to learning the community user's personality from the data provided by the community user, Personality settings 605 allow the system to set the community user's personality (e.g., passionate, sweet, lively, charming, etc.).

[0041] In addition to learning the expertise of community users through data provided by community users, the user interface 70 shown in Figure 7 can provide community users with expert descriptions 701, cultural roots 703, preferences 705, favorite foods 707, and family and relationships 709, allowing the customized community robot to more accurately match the attributes of the community users. Here, cultural roots 703 can be used to give the community robot more human-like characteristics and teach it how to generate content based on those settings. In this way, a user characteristic dataset describing community users can be formed from text formed by the relevant language model (step S305), providing a machine learning method for learning personal characteristics about community users and helping to create a community robot dedicated to each community user.

[0042] Furthermore, when the machine learning module running on the cloud server receives various data provided by community users, it performs preprocessing on the data (including processing the data with a language model and then forming a training dataset) (step S303). This allows the machine learning method to utilize the training dataset to train and create a community robot specifically for the community users (step S307).

[0043] Subsequently, the trained community robot is deployed to the community (step S309), and the collected data is used to establish a link with the community user's personal account, allowing the community robot to operate under the community user's personal account within the community (step S311). According to the embodiment, once the community robot is created, it can become a robot equipped with specialized knowledge and capable of utilizing specialized content. It can then be introduced into a company's webpage, any community, channel, blog, etc., to respond with content on behalf of community users or participate in community activities.

[0044] Furthermore, when community robots are deployed to a community and the corresponding community robot model is operated by a cloud server, the community robot will have an image of the community user (or an image set by the community user), and will be able to autonomously reply to community messages and interact with other community robots or real community users (for example, accepting friend requests, reading content and pressing the like button, collecting, sharing, commenting, and other community activities). In addition, it may autonomously generate content within the community that matches the individual's attributes and basic data using artificial intelligence, or generate appropriate content through interaction, and the content may be multimedia content such as text, graphics, audio, or video.

[0045] What is particularly noteworthy here is that the community robots generated through the customization provided by this invention, through Generative Artificial Intelligence (JAI) on a cloud server learning a large amount of data generated by various community activities using machine learning algorithms, not only possess the ability to generate characteristics and content of community users, but can also learn content in general within a community, much like humans do.

[0046] After deploying the community robot to the community, community users can operate the community's functions through a user interface launched by an application program. One such interface is the profile page 80 shown in Figure 8, which can display images of real community users and the community robot. It can also display information and functions of community users within the community, schematically showing friends 811, search 813, and account 815. Furthermore, it displays the community image 800 of the community robot, which may include simulated images of real people and community robots generated by the community robot model (for example, the first community robot 801, the second community robot 802, the third community robot 803, etc. shown in the figure, and community friends 804 formed by real people).

[0047] Next, as shown in Figure 9, a schematic diagram of an embodiment relating to a user interface utilizing a community user robot is shown. In the figure, a community robot top page 90, which is exclusive to community users, is shown. Here, an image 901 of the community robot (i.e., a simulated image of the community robot generated using the generative artificial intelligence described in the above embodiment) is shown, and various community activity data (including the number of chat messages 911, the number of shares 912, the number of likes 913, and functions such as collections 914 and shares 915) can be displayed on the image 901 of the community robot.

[0048] Next, Figure 10 shows another embodiment of a user interface utilizing a community user robot, in which a community robot dedicated to a community user can take on the role of an instructor or customer service representative with a specific professional background and interacts with other community users via an interaction page 110, which includes a first-stage dialogue 111 performed via text, in which the community robot model running on a cloud server acquires the meaning of words in the dialogue using a language model during the dialogue process, appropriately generates or shares video content 113 via the community robot, and further includes a second-stage dialogue 115.

[0049] In short, according to the above-described embodiment of the method, system, and non-temporary computer-readable media for customizing community user robots, the community robot operating within a community created by the method for customizing community user robots is a robot capable of promoting products or providing consulting customer service. The community robot can not only participate in community activities and interact with others in general, just like a real person, but can also recommend content or solutions depending on the chat target, and provide services as needed.

[0050] For example, a community user themselves can be an influencer (network influencer), and by customizing a community robot that has learned their expertise and abilities, it can become a representative of the influencer, responding to community users' messages quickly in real time, and differing from traditional influencers who can only perform limited community activities within a limited time. Influencers can upload their channel (URL), use a neural network to perform machine learning on the system, and train the system on the text and video content of the channel to create a community robot that will act on their behalf. From then on, the community robot can publish content on behalf of the influencer, respond to messages from internet users, and provide an even more complete service, eliminating reliance solely on the influencer or staff.

[0051] Furthermore, taking corporate users as an example, unlike individual community robots, when a company provides data on its related products, expertise, etc., the system can utilize machine learning algorithms to learn and then create a dedicated community robot for the company, which can then function as a smart customer service robot within the company's community.

[0052] The information disclosed above represents only preferred and implementable embodiments of the present invention, and the claims of the present invention are not limited thereto. Therefore, any equivalent technical modifications made using the description and drawings of the present invention are all included within the scope of the claims of the present invention. [Explanation of Symbols]

[0053] 10...Users 100...User devices 101...AI Robot 103...User image 105...Photography lens 200... Network 20...Cloud Server 21... Processing circuit 23...Memory 211...Natural Language Processing Module 211a...Language model 211b...Voice Model 212...Instruction Processing Module 213...User Interface Module 214... Machine Learning Module 214a...Generative Artificial Intelligence 214b...AI Robot Model 215...Database Module 25... Database 251...User Data 253... Artificial intelligence simulating human model 216...Community Module 201, 202, 203... User devices 40...User Interface 401...Option settings 403... Prompt 405...User Appearance 407...Prompt to generate appearance 50...Cerebrum 501...Web link 503... Document Upload 505...Personal community channel 60...User Interface 601...Availability settings 603...Audio settings 605... Personality settings 70...User Interface 701...Specialized description 703...Cultural Roots 705...Preferences 707... Favorite food 709...Family and Relationships 80...Profile page 811...friends 813... Search 815...account 800...Community images of community robots 801...The first community robot 802...Second Community Robot 803...The third community robot 804... Community friends 90...Community Robot Homepage 901...Image of a community robot 911... Chat message 912... Spread 913... High rating (likes) 914...Collection 915...share 110...Community Page 111...First paragraph dialogue 113...Video Content 115...Second paragraph of dialogue S301~S311: Process

Claims

1. A method for customizing a community user robot, which is executed on a cloud server, wherein the cloud server operates a community, and the method for customizing the community user robot is: The process includes receiving data provided by community users, performing preprocessing on the data, and then forming a training dataset after processing the data using a language model. The process involves using a machine learning method to train the aforementioned training dataset and create a community robot specifically for the aforementioned community users, This includes deploying the community robot to the community, creating a link with the community user's personal account, and operating the community robot under the community user's personal account in the community. How to customize community user robots.

2. The community users upload the data via a user interface launched in the application program, and the types of data include documents, graphics, video content, and website or channel links. A method for customizing the community user robot described in claim 1.

3. The user interface is for allowing community users to upload the data, and is also used to configure the style and profile of the community robot, provide personal data, configure community activities, and perform back-office management. A method for customizing the community user robot described in claim 2.

4. The user interface receives prompts used for setting the style, profile, and community activities of the community robot. A method for customizing the community user robot described in claim 3.

5. The data is personal data relating to the community user, and after processing with the language model, text is obtained for training the machine learning method to form a user feature dataset describing the community user. The machine learning method is provided for learning the personal features relating to the community user to create the community robot dedicated to the community user. A method for customizing the community user robot described in claim 2.

6. The aforementioned data includes video data of the community users, and combined with the personal characteristics of the community users obtained through learning, a generative artificial intelligence generates a simulated video of the community robot. A method for customizing the community user robot described in claim 5.

7. The community robot, which is exclusively for the community user, operates automatically within the community and processes the community user's community activities within the community. A method for customizing a community user robot according to any one of claims 1 to 6.

8. This includes the cloud server that runs the community. The method for customizing the community user robot is executed on the aforementioned cloud server, The method for customizing the aforementioned community user robot is: The process includes receiving data provided by community users, performing preprocessing on the data, and then forming a training dataset after processing the data using a language model. The process involves using a machine learning method to train the aforementioned training dataset and create a community robot specifically for the aforementioned community users, This includes deploying the community robot to the community, creating a link with the community user's personal account, and operating the community robot under the community user's personal account in the community. A system for customizing community user robots.

9. The community users upload the data via a user interface launched in the application program, and the types of data include documents, graphics, video content, and website or channel links, which are used to style the community robot, set up profiles, provide personal data, configure community activities, and perform back-office management. A system for customizing the community user robot described in claim 8.

10. The system for customizing the community user robot receives prompts via the user interface for setting the style, profile, and community activities of the community robot. A system for customizing the community user robot described in claim 9.

11. The data is personal data relating to the community user, which is processed by the language model to obtain text for training the machine learning method, forming a user feature dataset describing the community user, and the machine learning method is provided to learn the personal features relating to the community user, thereby creating the community robot dedicated to the community user. A system for customizing the community user robot described in claim 8.

12. The aforementioned data includes video data of the community users, and combined with the personal characteristics of the community users obtained through learning, a generative artificial intelligence generates a simulated video of the community robot. A system for customizing the community user robot described in claim 11.

13. The community robot, which is exclusively for the community user, operates automatically within the community and processes the community user's community activities within the community. A system for customizing a community user robot according to any one of claims 8 to 12.

14. In the aforementioned community, the community robot is operated via the cloud server, and the community robot is made to build community relationships with multiple users within the community and with each user's dedicated community robot, while carrying out community activities. A system for customizing the community user robot described in claim 13.

15. The program set is saved, the program set is executed on the user device's processor to launch an application program, and the application program connects to the cloud server in the system for customizing the community user robot described in claim 8 to cause the community user to create the community robot exclusively for them. Non-temporary, computer-readable media.

Images