KR102604759B1 - Method and system for providing community service based on metaverse - Google Patents

Abstract

A method of providing a community in a metaverse environment according to an embodiment for realizing the purpose of the present invention described above, comprising the steps of analyzing user-specific tendency information according to predetermined items from collected user information, at least one of the predetermined items Grouping users based on user preference information for each item, managing a list of interests for each community based on interest information of users in a community matching at least one group, based on the list of interests for each community , allocating a section for each interest on a map screen for each community in a metaverse environment, and reorganizing the section for each interest according to changes in the interest list.

Description

Method and system for providing metaverse-based community services {METHOD AND SYSTEM FOR PROVIDING COMMUNITY SERVICE BASED ON METAVERSE}

The present invention relates to a method and system for providing a metaverse-based community service. More specifically, the present invention relates to a method and system for providing a metaverse-based community service. More specifically, the present invention relates to a method and system for providing a metaverse-based community service, by analyzing the tendencies of users accessing the metaverse, grouping users, and analyzing the metaverse based on the interests of the grouped users. This relates to a method and system for providing a metaverse-based community service for managing sections of the community space implemented within the bus.

Metaverse is a compound word of the English word 'Meta', meaning 'virtual' or 'transcendence', and 'Universe', meaning universe, and refers to a virtual world where social activities similar to the real world take place.

Metaverse is a concept that is one step more advanced than virtual reality (a cutting-edge technology that allows people to experience life-like experiences in a virtual world created on a computer), and is not limited to just enjoying games or virtual reality using an avatar. It has the characteristic of being able to engage in social and cultural activities among members (avatars) similar to actual reality.

Meanwhile, virtual reality refers to an interface between humans and computers that creates a specific environment or situation with a computer, making it as if the person using it is interacting with the actual surrounding situation or environment.

This virtual reality is mainly intended to allow people to display and manipulate environments that are difficult to experience on a daily basis as if they were in the environment without directly experiencing them. Application areas include games, education, advanced programming, remote control, and remote satellites. These include surface exploration, exploration data analysis, and scientific visualization.

As a result, the metaverse can be seen as creating a specific environment or situation using a computer and allowing social ties between users to be more active within this environment or situation. In other words, users form various communities in the metaverse space with avatars representing themselves, and communicate among users with common interests or environments in the metaverse space. In other words, the communities implemented within the metaverse have common interests among users serving as their focal point.

Therefore, there is a need to support users to actively share common interests within the metaverse space, and since the metaverse space is a space where users move or engage in activities with their avatars, users with the same tendency can more effectively share common interests within a certain space. A method is required to allow interests to be reflected in the space so that interests can be shared and communicated.

One purpose of the present invention is to provide a metaverse-based community service, providing a platform to connect users, grouping users with the same tendency, and managing a plurality of communities grouped by tendency.

In addition, one object of the present invention is to create at least one section according to common interests on a map screen where an avatar can move or be active so that interests are reflected in the space according to the inflow or outflow of users with the same tendency for each community, and to create at least one section according to common interests. The purpose is to manage the size of the section according to the ratio of sharing users.

A method of providing a community in a metaverse environment according to an embodiment for realizing the purpose of the present invention described above, comprising the steps of analyzing user-specific tendency information according to predetermined items from collected user information, at least one of the predetermined items Grouping users based on user preference information for each item, managing a list of interests for each community based on interest information of users in a community matching at least one group, based on the list of interests for each community , allocating a section for each interest on a map screen for each community in a metaverse environment, and reorganizing the section for each interest according to changes in the interest list.

In one embodiment of the present invention, big data of analyzed user-specific tendency information and interest information for each user of a plurality of users in the community are analyzed to extract at least one interest information that is correlated to each community, and the community The star interest list may be based on at least one or more relevant interest information.

In one embodiment of the present invention, the interest list for each community may be changed according to user interest information flowing in and out for each community.

In one embodiment of the present invention, the list of interests for each community may be changed according to changes in interest information of users for each community.

In one embodiment of the present invention, the interest list for each community may include ranking information for each interest based on the ratio information of users for each interest.

In one embodiment of the present invention, the size or location of each interest section may be determined based on the interest ranking information.

In one embodiment of the present invention, the interest-specific sections may be allocated to sections pre-divided by ranking based on the interest-specific ranking information.

In one embodiment of the present invention, the predetermined item includes at least one of personality, taste, aptitude, method, expertise, preference, spending power, activity, and area of interest related to the user's inclination, and information for each item. The user's tendency can be analyzed by a combination of .

In one embodiment of the present invention, the step of grouping the users may be performed based on preference information for each item ranging from first priority to a predetermined priority according to the priority of each user.

In one embodiment of the present invention, the step of grouping the users includes, when a large number of users are concentrated in a specific group, adding the tendency information of the item with the next highest priority in the predetermined ranking to group the users included in the specific group. Can be regrouped.

According to an embodiment of the present invention, by grouping users with the same tendency and providing a plurality of metaverse-based communities in which users are grouped by tendency, users with the same tendency can easily share common interests, and , Although they are not currently interested, the interests of other users with the same tendency are managed as potential interests, so that users can not only have one interest, but also a number of common interests including current and potential interests, a metaverse-based community. Information can be shared and communicated within the space.

According to an embodiment of the present invention, various personal data such as each user's age, taste, personality, and interests are collected, users with high similarity are grouped, and information registered as interests by users in the group is collected. It is possible to provide a metaverse space optimized for groups grouped on the basis of common interests on a map screen where avatars can move or be active so that interests are reflected in the metaverse space according to the inflow or outflow of users with the same tendency for each community. By creating at least one section accordingly and managing the size of the section according to the ratio of users sharing the interest, it is possible to support users to effectively access the space where their interests are projected in the metaverse space.

Figure 1 is a block diagram for explaining a community system according to an embodiment of the present invention.
FIG. 2 is a block diagram for explaining the central server included in the system of FIG. 1.
FIG. 3 is a block diagram for explaining the community management unit of FIG. 2.
FIG. 4 is a block diagram for explaining the map screen management unit of FIG. 2.
Figure 5 is an example diagram for explaining a map screen in a metaverse environment.
Figure 6 is a flowchart illustrating a method for providing a community according to an embodiment of the present invention.

Hereinafter, embodiments disclosed in the present specification will be described in detail with reference to the attached drawings. However, identical or similar components will be assigned the same reference numbers regardless of reference numerals, and duplicate descriptions thereof will be omitted. The suffixes “module” and “part” for components used in the following description are given or used interchangeably only for the ease of preparing the specification, and do not have distinct meanings or roles in themselves. Additionally, in describing the embodiments disclosed in this specification, if it is determined that detailed descriptions of related known technologies may obscure the gist of the embodiments disclosed in this specification, the detailed descriptions will be omitted. In addition, the attached drawings are only for easy understanding of the embodiments disclosed in this specification, and the technical idea disclosed in this specification is not limited by the attached drawings, and all changes included in the spirit and technical scope of the present invention are not limited. , should be understood to include equivalents or substitutes.

Terms containing ordinal numbers, such as first, second, etc., may be used to describe various components, but the components are not limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.

When a component is said to be "connected" or "connected" to another component, it is understood that it may be directly connected to or connected to the other component, but that other components may exist in between. It should be. On the other hand, when it is mentioned that a component is “directly connected” or “directly connected” to another component, it should be understood that there are no other components in between.

In this application, terms such as “comprise” or “have” are intended to designate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, but are not intended to indicate the presence of one or more other features. It should be understood that this does not exclude in advance the possibility of the existence or addition of elements, numbers, steps, operations, components, parts, or combinations thereof.

In this specification, 'part' includes a unit realized by hardware, a unit realized by software, and a unit realized using both. Additionally, one unit may be realized using two or more pieces of hardware, and two or more units may be realized using one piece of hardware.

In this specification, some of the operations or functions described as being performed by a terminal, apparatus, or device may instead be performed on a server connected to the terminal, apparatus, or device. Likewise, some of the operations or functions described as being performed by the server may also be performed in a terminal, apparatus, or device connected to the server.

The present invention can be implemented as computer-readable code on a computer-readable recording medium, and the computer-readable recording medium includes all types of recording devices that store data that can be read by a computer system. . Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage devices. Additionally, the computer-readable recording medium can be distributed across computer systems connected to a network, so that computer-readable code can be stored and executed in a distributed manner.

Hereinafter, the present invention will be described in detail with reference to the attached drawings.

1 is a diagram for explaining a community system according to an embodiment of the present invention. Referring to FIG. 1, the metaverse-based community system 1000 according to this embodiment connects at least one user terminal 100 and at least one user terminal 100 who want to participate in the community of the metaverse environment. and may include a central server 300 that creates and manages a metaverse-based community. However, since the community system 1000 of FIG. 1 is only one embodiment of the present invention, the present invention is not limitedly interpreted through FIG. 1.

For example, each component in Figure 1 is generally connected through a network (network, 10). That is, as shown in FIG. 1, the at least one user terminal 100 may be connected to the central server 300 through the network 10. In addition, the community system 1000 according to the present embodiment, separately from the user terminal 100, includes an input/output device 150 for outputting a screen implemented in a metaverse environment and receiving user manipulation in the metaverse environment. It can be included. The input/output device 150 can be connected to the user terminal 100 by wired or wireless communication, and can connect to the central server 300 through the user terminal 100 to access the metaverse environment. However, it is not limited to this. The input/output device 150 may be directly connected to the central server 300 through a network without the mediation of the user terminal 100. Alternatively, the user terminal 100 itself may be an input/output device 150. Meanwhile, the input/output device 150 may be composed of separate output devices and input devices. In addition, the community system 1000 may further include an external information server that transmits online personal data to analyze the user's interests.

Here, the network refers to a connection structure that allows information exchange between nodes such as a plurality of terminals and servers. Examples of such networks include RF, 3rd Generation Partnership Project (3GPP) network, and Long Term (LTE) network. Evolution) network, 5GPP (5th Generation Partnership Project) network, WIMAX (World Interoperability for Microwave Access) network, Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network) , PAN (Personal Area Network), Bluetooth (Bluetooth) network, NFC network, satellite broadcasting network, analog broadcasting network, DMB (Digital Multimedia Broadcasting) network, etc., but are not limited thereto.

In the following, the term 'at least one' is defined as a term including singular and plural, and even if the term 'at least one' does not exist, each component may exist in singular or plural, and may mean singular or plural. It is self-evident that it exists. In addition, whether each component is provided in singular or plural form may be changed depending on the embodiment.

At least one user terminal 100 uses community service-related web pages, app pages, programs, or applications to display characteristics, tastes, aptitudes, methods, expertise, preferences, spending power, activity, and areas of interest related to the user's inclinations. It may be a user terminal that extracts or transmits tendency information, which is a variety of personal data, and outputs a customized community suitable for each individual based on this information. In addition, in addition to the propensity information, the user terminal 100 provides general customer information, such as name, ID, address, etc., and gender, age, physical condition, and income corresponding to the user's environmental information. , SNS activities, religion, educational background information, etc. can be received and managed. However, it is not limited to this. The information is matched for each user and managed as user information, and can be used to group users.

The user terminal 100 is connected to the central server 300 and supports the user to conduct a community in a metaverse environment. For example, it is configured to support access to the metaverse community in a 2D environment through the user terminal 100, or access to the metaverse community in a 3D environment through the input/output device 150, which will be described later.

In other words, the community system 1000 according to the present embodiment manages a list of interests of users for each group based on the user's personal data, and not only matches the interests registered or extracted with the users in the group. In addition, it supports information sharing and communication about various interests within the community by matching interests registered or extracted by other users with potential interests.

Here, at least one user terminal 100 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, a laptop equipped with a navigation system and a web browser, a desktop, a laptop, etc. At this time, at least one user terminal 100 may be implemented as a terminal capable of accessing a remote server or terminal through a network. At least one user terminal 100 is, for example, a wireless communication device that guarantees portability and mobility, and includes navigation, personal communication system (PCS), global system for mobile communications (GSM), personal digital cellular (PDC), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet) ) It may include all types of handheld-based wireless communication devices such as terminals, smartphones, smartpads, and tablet PCs.

Meanwhile, the input/output device can be used to access the metaverse community in a 3D environment. The input/output device 150 can be connected to the user terminal 100 by wired or wireless communication, and is configured to connect to the central server 300 through the user terminal 100 and access the metaverse environment. For example, the input/output device 150 may be comprised of an output device and an input device.

First, the output device is a device that outputs a screen implemented in a metaverse environment and can be electrically connected to the user terminal 100.

Here, metaverse is a compound word of meta, meaning processing and abstraction, and universe, meaning the real world. In an online three-dimensional virtual world, individuals embodied as characters or avatars communicate with each other, It refers to a platform that allows you to carry out activities in the real world. This metaverse environment may include extended reality (XR), which encompasses virtual reality (VR), augmented reality (AR), and mixed reality (MR).

Virtual reality refers to a technology that uses a computer to build a virtual space that does not exist in the real world and then makes the virtual space feel like reality, while augmented reality or mixed reality is a technology that adds computer-generated information to the real world. Technology, that is, technology that allows real-time interaction with users by combining the real world and the virtual world. The metaverse environment in the present invention can be implemented through any one of virtual reality, augmented reality, and mixed reality. You can.

The output device is worn by the user to output images on the screen implemented in the metaverse environment. For example, a display device that is mounted on the user's head or eyes and outputs VR/AR images directly in front of the user's eyes. It can be provided as a head mounted display (HMD), etc. A head-mounted display is a type of wearable video output terminal that allows users to experience realistic images. The display is mainly mounted on the front of the helmet and can output images such as VR/AR.

The output device can track the user's gaze through the head movement of the user wearing it, and can process the image to be changed and displayed according to the user's gaze movement through head tracking. In other words, the output device can quantify the rotation angle and speed of the user's head and process the image so that it changes in the same direction as the movement of the user's head. For example, if the user's head turns to the right, the image on the screen implemented in the metaverse environment will move to the right, and if the user's head rotates to the left, the image on the screen implemented in the metaverse environment will also move to the left. You can. That is, the image of the output device includes a community object, and the user accesses information of interest by inputting various operations through an input device described later.

The input device is a device worn by the user and generates a manipulation signal for a command according to the user's manipulation, and may be electrically connected to the output device or the user terminal 100.

The input device is worn on the user's hand to interact on a screen implemented in a metaverse environment, for example, VR that tracks the user's motion, provides physical feedback, and provides tactile feedback. /Can be provided with AR gloves (VR/AR gloves).

These VR/AR gloves track the motion of the user's fingers, hands, and wrists and implement them in VR/AR, and apply a preset amount of tension to each finger to improve the hardness of objects in VR/AR. The texture is expressed through the vibration sensor installed on each finger. In the present invention, when the user generates a manipulation signal using a VR/AR glove to interact in the metaverse environment, the manipulation signal is used to control the metaverse-based Community services, such as allowing access to the space, may be provided.

However, it is not limited to this. For example, depending on the type of input/output device, the user may not wear a separate input device, and the camera tracks the positions of both hands of the user outside of the head-mounted display worn as the output device and converts the corresponding manipulation signals into input signals. It can also be converted and provided. Additionally, the presence or absence of a haptic function that provides a sense of touch may be determined depending on the type of input/output device 150. In addition, the input/output device 150 can be directly connected to the central server 300 through a network without the mediation of the user terminal 100. Alternatively, the user terminal 100 itself may be an input/output device 150.

The central server 300 may be a server that provides a community service web page, app page, program, or application. In addition, the central server 300 is related to the user's tendency input from the user terminal 100 or an information provision server, or analyzed from user log information such as clicks and residence time within a community service web page, app page, etc. It may be a server that clusters users with a certain degree of similarity based on disposition information, which is a variety of personal data such as personality, taste, aptitude, method, expertise, preference, spending power, activity, and area of interest. For example, user-specific tendency information can be quantified and output as a scatterplot on a graph, and users can be clustered based on the clustered scatterplot. Meanwhile, when outputting a scatterplot, it may be based on demographic data provided from an external server.

Additionally, the central server 300 may be a server that collects the user's online interest search history, interest information issues, and shopping history provided from the user terminal 100 or the information provision server and analyzes the user's interests. Additionally, the central server 300 may be a server that extracts an interest list of a community of users grouped as a result of analysis, rearranges and manages the list based on the ratio of users sharing each interest. In addition, the central server 300 may be a server that collects information such as individual profiles and community interests to match user-specific interests, builds big data, and performs training and learning.

Here, the central server 300 may be implemented as a computer that can connect to a remote server or terminal through a network. Here, the computer may include, for example, a laptop equipped with a navigation system and a web browser, a desktop, a laptop, etc.

FIG. 2 is a block diagram for explaining the central server included in the system of FIG. 1. FIG. 3 is a block diagram for explaining the community management unit of FIG. 2. FIG. 4 is a block diagram for explaining the map screen management unit of FIG. 2.

2 and 4, the central server 300 includes a collection unit 310, an analysis unit 320, an extraction unit 330, a user grouping unit 340, a community management unit 350, and a map screen management unit. It includes 360 and an output unit 370.

The central server 300 according to an embodiment of the present invention or another server (not shown) operating in conjunction with the user terminal 100, at least one user terminal 100, and at least one information providing server provide a community service. When transmitting an application, program, app page, web page, etc., at least one user terminal 100 and at least one information provision server may install or open a community service application, program, app page, web page, etc. there is. Additionally, a service program may be run on at least one user terminal 100 and at least one information provision server using a script executed in a web browser. Here, a web browser is a program that allows the use of web (WWW: world wide web) services and refers to a program that receives and displays hypertext written in HTML (hypertext mark-up language). For example, Netscape, Includes Explorer, Chrome, etc. Additionally, an application refers to an application on a terminal and includes, for example, an app running on a mobile terminal (smartphone).

The collection unit 310 may collect user information, such as collecting personal information and interest information registered from the user terminal 100. The user information is used as information for identification of each user or as basic data for analyzing user tendencies, and is used to analyze user tendencies for each user in the analysis unit 320, which will be described later.

The user information is provided through the user terminal 100 to general customer information such as name, ID, address, etc., and gender, age, physical condition, income, SNS activity, religion, etc. corresponding to matching information for user targeting. Includes educational background information, etc., and includes personality information, which is personal data in various fields such as personality, taste, aptitude, method, expertise, preference, spending power, activity, and area of interest related to the user's community tendency for user grouping. do. The information includes information directly input from the user and information calculated from information input from the user.

Propensity information may be provided by user input for each item, or may be extracted from other information. For example, the user's area of interest can be extracted by collecting online interest search history, interest information issues, and shopping history provided from the user terminal 100 or the information provision server for each user. This information is personal information, and consent to provision of information may be provided. In addition, the collection unit 310 is configured to collect visit history of sections on which interests of each user are projected on the map screen within the community of each user managed by the central server 300. The visit history information is managed as user information.

Based on the user information collected by the collection unit 310, users can be grouped into a new group or matched to an existing group. Meanwhile, the user information is used as basic data for analyzing the user's tendency, and the analysis unit 320, which will be described later, extracts the user's tendency information, and accordingly, the user grouping unit 340, which will be described later, It is configured to group. For example, in order to extract user's tendency information, information can be collected from questionnaire methods, that is, results extracted from answers to multiple questions. Additionally, the user's personality information may be based on the results of MBTI, or may be based on the results of MBTI for interests prepared in association with community interests.

Additionally, at least one piece of information may be collected through visit history for each section on which interests are projected on the map screen for each user. For example, visit interest information, interest field, type, visit time, frequency information, community access time, output and operation information based on log information of the user's input/output device, product information purchased within the community, user input Feedback information, etc. may be collected. The information can be extracted as common information about the interests of users belonging to each group within the community.

The analysis unit 320 is configured to analyze the user's tendency based on the user information. That is, as described above, it is configured to analyze tendency information for each user based on information through multiple routes, such as information input from the user, information extracted from information input from the user, and information provided from an external server. That is, the analysis unit 320 is configured to analyze the user's tendency based on information collected for at least one item. For example, by item consisting of at least one or more personal data such as personality, taste, aptitude, method, expertise, preference, spending power, activity, and area of interest related to the user's community tendency, the user It is structured to identify the tendencies and priorities of For example, it is a method of identifying tendency information such as priority by directly entering the ranking of each item, or extracting common information based on at least one piece of information based on the user's search, click, detailed information stay time information, etc. It may include a method of identifying tendency information including user interest or priority information.

The analysis unit 320 may analyze user-specific tendency information through a combination of at least one item-specific information. For example, the analysis unit 320 may analyze the user's tendency information by assigning weight to each information collected through multiple routes. In the case of the weight, weight may be given to information directly input by the user, or if the same information is extracted through multiple routes, weight may be given to the information. The propensity information is an indicator for analyzing the propensity of each user, and can be analyzed into items such as personality, taste, aptitude, method, expertise, preference, spending power, activity, and area of interest related to the user's propensity, which is the personalized data described above. You can. The item may be subdivided into sub-items. For example, spending power can be subdivided according to section, and in the case of interest area items, it can be subdivided into major categories, medium categories, and small categories to which the interests belong. Additionally, each item may be subdivided into specific items or sections. However, it is not limited to this.

Additionally, the information for each item may be expressed as specific information or numerical information. For example, when a user searches for a specific community in the search history, or based on the information entered when registering as a member of the service, user preference information can be extracted based on representative tendencies or interest information in the community. You can. Of course, the relevant information may be represented as at least one piece of information according to priority. For example, in the field of exercise, it may be specified whether the user prefers to directly exercise or only watch sports broadcasts. Additionally, specific information among each item of information may be converted into numerical information with predetermined labeling and recorded. For example, in preference items, numerical information pre-specified for each field, for example, '1' may be assigned to exercise, and a numerical value of 'n' may be assigned to music. However, it is not limited to this. In addition, additional quantification can be done according to the section for each detailed item, and the tendency for each subdivided item can be derived through the listed numbers.

That is, the analysis unit 320 may classify users into similar groups in the user grouping unit described later according to the user's tendency information combined with the item-specific information. For example, group classification may be based on a combination of 2-dimensional, 3-dimensional, or N-dimensional items depending on the number of items. For example, a classification combining preference item information and interest area item information may be applied. Meanwhile, an item may have sub-items depending on the information. Of course, each sub-item can also be subdivided into additional sub-items.

Meanwhile, the analysis unit 320 may perform cohort analysis based on the user information. For example, the analysis unit 320 may perform a cohort analysis of each user's tendency based on user information for each user, and may analyze each user's tendency through a combination of at least one cohort item. In other words, each item for each propensity can be treated as a cohort item.

For example, the analysis unit 320 may classify user information according to a cohort classification combining the cohort items. For example, the cohort classification may represent information that combines two-dimensional, three-dimensional, or N-dimensional cohort items depending on the number of cohort items. For example, cohort classification combining age information and preference information may be applied. Meanwhile, a cohort item may have sub-cohort items depending on the information. For example, the age cohort item may consist of Cohort 1 for users in their teens, Cohort 2 for users in their 20s, Cohort 3 for users in their 30s, etc. Of course, each Cohort 1, Cohort 2, and Cohort 3 can be subdivided into subcohorts such as Cohort 1a, Cohort 1b, and Cohort 1c according to detailed classification criteria (e.g., early, mid, and late) within each cohort. You can. Likewise, Cohort 1a can also be subdivided again. In other words, when analyzing user tendencies according to N cohort items, each user can be classified as having tendencies classified by N cohorts. Meanwhile, priority may also be given to cohort items. For example, users can be output as a scatterplot based on some cohort items according to priority among multiple cohort items. That is, in the user grouping unit 340, which will be described later, users can be placed on the graph according to the priorities of the cohort items, and when it is necessary to divide users into groups, users can be further classified according to the next-highest priority cohort items.

Meanwhile, in order to obtain user information, the questionnaire method can be applied, and necessary data can be extracted from answers in subjective form as well as multiple choice form. Accordingly, in order to analyze the user's tendency information based on the input information, text analysis technology based on artificial intelligence can be applied. For example, when learning a model and model for classifying the described text into specific information for each item, the database described above can be used as a training dataset. Accordingly, one embodiment of the present invention is an NLP preprocessor that receives data text from a database as input and analyzes it in units of morphemes to extract only meaningful information in propensity analysis, and all morphemes extracted from one sentence are classified according to predetermined standards. It may further include a training data generator that labels with information, and a CNN or RNN model that learns the generated training data.

First, natural language processing must be performed. To this end, in one embodiment of the present invention, the input text collected for learning is composed of Korean words or sentences. Hangul has a complex language structure with multiple parts of speech mixed together, so word expressions must be created appropriately through NLP steps. Therefore, after receiving a text file as input and analyzing it in units of morphemes using a morpheme analyzer, only morphemes corresponding to preset tags can be extracted. Next, learning data must be created, which can be done by labeling the morpheme bundles extracted from the Korean natural language processor as a single piece of information according to predetermined classification criteria. Next, the CNN model converts the words in the sentence into low-dimensional vectors and then performs convolution with the converted word vectors based on filters of various sizes. The results obtained through this transformation can be classified into a single feature vector through the process of max pooling and dropout normalization.

At this time, the text of the database used can be in a format where the collection of keywords and labels that have gone through the input preprocessing process are separated by commas. In order to use such data as input to a CNN model, a word processing process may be additionally performed to convert a collection of keywords into a list and store it with a label. By constructing the first layer by embedding the index of the data created in this way into a low-dimensional vector, a convolution operation is performed on the formed low-dimensional vectors through filters of different sizes, and a maximum of merging them into one large feature vector is performed. A pooling process may be performed. At this time, nonlinearity can be applied to the result of performing the convolution operation, and the sentence can be slid without edge padding through VALID (effective value) padding. Afterwards, overfitting of the CNN can be prevented through a dropout regularization layer, and the layer can stochastically deactivate some of the neurons, preventing mutual adaptation of neurons and forcing them to learn features individually. Finally, the label with the highest score according to matrix multiplication through the feature vector can be selected and predicted. Based on the final score converted to normalized probability, loss and accuracy can be calculated using cross-entropy loss, a standard loss function for classification problems, and accuracy can be measured based on this.

The extraction unit 330 analyzes the big data of the analyzed user-specific preference information and interest information for each user, and extracts the correlation between the preference information of the plurality of users and the interest information for each user. It is composed. That is, the extraction unit 330 is configured to extract correlations for each piece of information from information for each item from the tendency information for a plurality of users and interest information input or extracted for each actual user.

That is, by analyzing big data of the analyzed user-specific tendency information and interest information for each user of a large number of users in the community, at least one interest information with correlation for each community is extracted, and the list of interests for each community is stored in the correlation chart. It can be configured based on at least one piece of interest information.

Accordingly, detailed information that influenced actual interests can be extracted according to each user's tendencies. For example, the correlation between a specific tendency and the sports field can be extracted through common information of a group of users who have a specific tendency and interest or priority in the “sports field.” For example, if a user with a specific tendency is interested in sports A, B, and C, sports that are their common interest information can be extracted as common information.

Meanwhile, the extraction unit for this embodiment is not limited to information for each item of the same nature in extracting the correlation between user-specific tendency information and interest information. For example, if there is a priority for a specific item of user-specific tendency information, it is limited to the specific item and common information is not extracted, and if a specific interest is common from multiple interest information, the interest related to the interest field item is not extracted. Common information can be calculated.

In other words, extract the correlation of common information for each detailed item focusing on items with high correlation with information about the community, including primarily related information, but also extract information on different items where the presence or absence of correlation is secondarily suspected. Try to extract correlation. In other words, the correlation between the tendency and at least one piece of unspecified information is extracted from the interest information of multiple users with the same tendency. The correlation may be determined by the frequency of each information. Among the interest information of multiple users with a specific tendency, information with a high frequency is extracted with high correlation, and in the case of a community with the combined information, if it contains a certain number, it is matched to the specific tendency. As a result, one community may be matched with several groups based on similarity.

Meanwhile, a high correlation or correlation of a specific item means that it is higher than that of other items or has a correlation or correlation above a certain value, which is a predetermined standard.

When an item with a high correlation with specific propensity information and specific information about the item are extracted from the extraction unit 330, a relationship diagram between the item, specific information about the item, and the specific propensity can be generated. In other words, when users are grouped by propensity by the user grouping unit described later based on propensity information, nodes are formed for each grouped propensity, and a relationship diagram in which each interest information is connected through a network as a node can be created. . For example, it may be possible to connect to a network only when nodes have a certain level of quantified correlation.

Meanwhile, the extraction unit 330 learns the correlation between each segmented item or information based on the tendency information and interest information of an unspecified number of people based on artificial intelligence, and creates a database of the correlation between the items and information, Each time data is updated, the correlation may be configured to update. Accordingly, based on an artificial intelligence-based correlation analysis model, a community containing interest information with a high correlation to the analyzed user tendency is recommended, or when a new community is registered, interest information is analyzed to determine a specific tendency. It can be matched to a user group with .

For example, artificial intelligence learning can be done in the form of unsupervised learning, or alternatively, partially supervised learning.

For example, in learning the correlation of information, generally published statistical data can be referred to, and for some of the information related to the user's tendency, statistical reports according to the user's preference by age can be referenced. Of course, if special information is included depending on the age and region of each user, supervised learning can be performed accordingly. Meanwhile, in extracting correlation, not only the correlation of each information unit but also the overall correlation of one cluster of data formed by each information can be extracted, and learning can be performed on this. In this case, if specific information belongs to the corresponding cluster data, it may be output as having a correlation based on it.

In addition, in the case of user-specific propensity information, if it is interest information that is directly related to the propensity-specific item itself, weight is given to the information itself, or the relationship between the propensity information and information analyzed as having a correlation By assigning a weighted sum to , a meaningful correlation analysis model can be created. For example, interest information may also be related to disposition information. For example, if one is interested in a specific sport, tendencies such as active or extroverted can be extracted depending on the sports personality.

Meanwhile, the correlation analysis model can be continuously updated. For example, in the case of an item that has a significant network correlation based on a random item (node), it is updated to a specific item and fed back to the next data learning cycle. In other words, in the next data learning cycle, the updated specific item is used to recalculate the correlation indicating the correlation with any item used as a standard, and the network connected to the specific item continuously identifies another meaningful item. It is configured to search. If there are no items with significant network correlation and no other specific items can be discovered, the discovery process is terminated. For example, in order to analyze meaningful network correlation, the minimum required amount of data is specified, and if a specific item of user information collected in the collection unit 310 meets the minimum required data amount, an update is performed. can do.

The user grouping unit 340 is configured to group users into groups composed of users with a certain degree of similarity based on the tendency information analyzed for each user. For example, based on information for each item, users who share at least one piece of information can be grouped into the same group. The user grouping unit 340 may group users based on preference information for each item ranging from first priority to a predetermined priority according to the priority of each user. Additionally, when a large number of users are concentrated in one group, the user grouping unit 340 may subdivide and regroup information for each item shared in the grouping. For example, when a large number of users are concentrated in a specific group, the users included in the specific group can be regrouped by adding the tendency information of the next item in the predetermined ranking. In other words, if the user group is biased to one side and it is difficult to classify it into a certain number of groups, the biased group is set as the analysis range and the users in that range are regrouped again. In addition, if additional regrouping is required for the regrouped user group, the regrouping can be done in the same manner.

Meanwhile, the user grouping unit 340 can apply various methods in regrouping users. For example, user groups may be initially grouped according to higher-level items, and if regrouping is required, the user groups can be regrouped by subdividing them into lower-level items. That is, in the case of users grouped with the same or similar tendencies based on the priority of the preference item, etc., they may be regrouped according to the priority of subordinate items of the preference item. Alternatively, in a user group represented by scattered points with n-dimensional coordinates according to the number of items or information shared by tendency information by the coordinate method, if regrouping is required, items or information are added to create an n + 1-dimensional By representing scattered points with coordinates, users can be regrouped based on the distance between the scattered points. For example, if regrouping is necessary for a user group grouped based on the user's personality and preferences, the group can be regrouped by adding interest items that are the next priority after the personality and preferences. Of course, if regrouping is required even after this, regrouping can be performed by adding items or information of the next rank and expressing them in n + 2-dimensional coordinates.

Meanwhile, at this time, the group concept may be a method of recommending a group similar to the user's tendency information or similar to the user's tendency by calculating similarity according to the group's tendency based on the group's profile. At this time, if the group's profile information is clear and has various propensity values, it may be implemented as a content-based group recommendation method rather than collaborative filtering. In other words, a group is not a fixed concept, and since it is grouped based on the tendency information of a plurality of users, it should be understood as a concept that can vary depending on the grouped users. This user group method is intended to recommend interests within the group to which each user belongs, taking into account the individual's tendencies and the tendencies of other users with the same tendencies, in order to ensure satisfaction for each user. Accordingly, information reflecting trends can be managed according to the influx of new users organized within the same group, and users belonging to the existing group can also be recommended more diverse interests as new users flow in.

First, in grouping, propensity information and the Euclidean calculation method can be used to calculate the average and standard deviation of the similarity of the propensity information for each user. At this time, the average and standard deviation can be used to calculate the central value for each individual's propensity, and multidimensional (propensity information) values can be calculated to calculate distance based on the central value for each propensity. Additionally, distance-based similarity between groups can be calculated. If the distance between groups is more than a certain level, they can be grouped into different groups, and if the distance between groups is less than a certain level, they can be grouped into the same group. Of course, various methods other than those described above may be used. At this time, the meaning of the above-mentioned group may mean a group of individuals with similar demographic tendencies.

In contrast, raw data related to the user's tendency information is collected from the user terminal or external server, and unstructured data, structured data, and semi-structured data contained in the raw data are refined, After preprocessing including classification with metadata, analysis including data mining can be performed on the preprocessed data. At this time, data mining involves classification, which predicts the class of new data by exploring the inherent relationships between preprocessed data and learning a training data set for the class, or clustering, which groups data based on similarity without class information. Clustering) can be performed. Of course, various other mining methods may exist, and mining may be done differently depending on the type of big data collected and stored or the type of query to be requested later. Big data constructed in this way can also go through a verification process using artificial neural network deep learning or machine learning.

The community management unit 350 is configured to manage information on at least one community and match communities for each group classified based on the user's tendency information. Additionally, the community management unit 350 is configured to manage the interests of users belonging to each community and reorganize the interest list according to the inflow or outflow of users within the community. For example, the community management unit 350 includes a community information management unit 351, an interest list management unit 353, and a list reconstruction unit 355.

The community information management unit 351 is configured to manage interest information of users belonging to at least one community collected through the collection unit 310. That is, in the user grouping unit 340, users may be grouped according to the preferences of each user, and a community may be matched for each group, or a plurality of groups may be matched into one community. At this time, the groups that are matched to one community are based on the premise that the user's tendency information or interest information belonging to each group has a degree of similarity above a certain level.

The interest information is community-level information and includes information about the interests of member users, for example, areas of interest, products of interest, degree of interest, etc. Meanwhile, the information about the interests may be subdivided into major categories, medium categories, and small categories to which the interests belong according to predetermined classification criteria. Additionally, the interest information includes information input from the user terminal 100 or information extracted based on correlation in the extraction unit of the central server 300.

That is, the community information management unit 351 includes interest information managed for each community, member information, for example, user information as a member, group information matched to the community, user information for each group, and community characteristic information, for example, community It is configured to manage map information and information for each section on the map.

The interest list management unit 353 selects the community in which the interest information with high correlation extracted from the extraction unit 320 is matched for each community to which the group classified based on the tendency information in the user grouping unit 340 is matched. It is configured to manage a list of interests.

That is, in a group classified based on user-specific tendency information, information with high correlation is extracted as community interest information from the user information of users belonging to the group, and the interest list management unit 353 determines the correlation for each community. It is configured to manage a high-match interest list containing high information. In other words, the interest list is matched for each community. For example, in the case of community A, the first interest list, and in the case of community B, the second interest list may be matched and managed.

Additionally, the interest list management unit 353 may generate and output a list of at least one interest according to the ranking according to whether or not it contains a lot of highly correlated information for each community. That is, the interest list for each community may include ranking information. For example, interests containing highly correlated information can be output for each community, and the rankings can be sorted according to the number of matches of highly correlated information. For example, according to the ranking in which multiple pieces of highly correlated information are matched, they may be arranged as 1. a interest, 2. b interest, 3. d interest, etc. In other words, a user belonging to a specific community can sign up for the service and receive a recommended list of interests at the time of access.

Additionally, the interest list management unit 353 is configured to manage interest information for each community based on interest information for each user. For example, if user ‘A’ is interested in a, b, and c within a specific community, and user ‘B’ is interested in a, d, and f, the interest list management unit 353 will: As interests of a specific community, a, b, c, d, and f can be managed as a list, and a can be managed by treating it as a priority as a common interest. At this time, d and f can be managed as potential interests for 'A', and a and b can be managed as potential interests for 'B'. Meanwhile, as interest information, it is managed as information according to major, medium, and small categories for each item of interest, and is configured to manage information according to community information changes, updates, and promotions in real time. For example, if a new subscriber, 'Sick', is grouped with the same tendency and matched to the same community, if 'Sick''s interests are a and k, k can be registered and managed as a new interest in the community. .

Additionally, the interest list management unit 353 may manage a similar interest list according to the similarity of each community based on tendency information. For example, according to the similarity based on the tendency information of each user belonging to each community, each matching interest list can be managed as a similar interest list. In the above example, when the preference information of users belonging to Community A and Community B have a high degree of similarity for each item, the first interest list and the second interest list may be managed as similar interest lists. The similarity for each group may be determined depending on whether the propensity information for each item matches. In addition, in the case of a user group represented by scattered points with n-dimensional coordinates according to the number of items or information shared for each tendency information by the coordinate method, the degree of similarity can be determined according to the proximity of the group shown on the coordinates.

Accordingly, it is possible to create a list of interests that primarily matches the community to which the user belongs, and if there is no object of interest tailored to the user in the list of interests, interest lists with high similarity are sequentially included, allowing the user to participate. We can support you to do so.

The list reconstruction unit 355 is configured to reconstruct the interest list for each community according to changes in interest information of users who are members of each community. For example, the list reconstruction unit 355 is configured to change the interest list of the community accordingly when a change occurs, such as adding or deleting a user's interests in the community. Changes in interests for each user include both new users being added to the community or changes in interests from existing users. In other words, the interest list for each community may be changed according to user interest information flowing in and out for each community, and may be changed according to changes in user interest information for each community.

For example, the list reconstruction unit 355 may manage ranking information of interests for each community according to changes in the user's interest information. For example, common interests input or extracted from multiple users can be managed by ranking according to the number of matching users. There may be changes in the ranking information, and the ranking information may be based on allocating interests for each section by the map screen management unit 360, which will be described later.

Meanwhile, the ranking of interests may be determined based on the number or ratio of users for each interest, and may be based on each interest and the user's activities, for example, the user's interest search history, section visit history, interest information issues, related interest shopping history, etc. Thus, you can determine the ranking by scoring each interest. For example, the interest information for each user is scored according to relevance based on at least one or more information calculated from the user information, such as the user's visit history, time spent in the section during the visit, log information recorded during the stay, etc. And by accumulating this, the ranking of interests for each community can be determined.

In addition, within each user's community service web page, app page, etc., specific interests can be scored based on tendencies such as user log information such as clicks and dwell time for specific interests. Meanwhile, the points allocated to each characteristic or behavior for scoring may be determined in advance. For example, 10 points may be given for purchasing a product, 1 point may be given for a click, etc.

The list reconstruction unit 355 regularly or irregularly collects interest information about the user's current status, and accordingly, not only public status information for a specific point in time, but also the change process of interest for each user over time. It is configured to collect and manage information of interest by accumulating information on human status. Accordingly, new interests may be added for each community, or if a specific interest is deleted from the user's interests, it may be deleted from interest information for each community. Meanwhile, the community system 1000 according to this embodiment is configured to allocate a section in which interests within the map screen are projected based on the interests of each community based on the metaverse. In other words, deletion of an interest may mean that the user has never visited the section where the user's interest is projected for a certain period of time. That is, if not all users in the community visit a specific section, interests assigned to that section may be deleted from the list, and at the same time, the assigned section may be deleted from the map screen.

That is, the community system 1000 according to this embodiment is configured to manage communities matched to each group based on the user's personal data and to manage a list of interests of users who are members of each community. In other words, each user's current interests and potential interests that are the interests of other users in the same community are managed, and changes in the user's interests are reflected on the map screen, allowing users to easily access their interests while active within the metaverse community. there is.

The map screen management unit 360 is configured to manage the map screen of a community based on interest information for each community managed by the community management unit 350. The map screen management unit 360 is configured to configure a map screen based on interest information for each community and update the map screen according to changes in interest information. The map screen management unit 360 may manage the map space with a 2D image as shown in FIG. 5, or alternatively, it may be configured to manage the map space with a 3D image. The map screen is configured to display a space where an avatar representing the user can move or be active, and as shown in FIG. 5, it includes a plaza where at least one or more users can gather or a passage through which they can move. It is structured to include areas where interests are projected for each section. The map screen management unit 360 may include an interest section allocation unit 361, a section management unit 363, and a section update unit 365.

The interest section allocation unit 361 is configured to allocate a section in which the user's interests are projected within the map screen for each community. The map screen is an area where users of each community can move or engage in activities in the metaverse space, assuming that it is a limited map space.

That is, the interest section allocator 361 is configured to allocate sections for each interest to a certain area so that users can visit them in a limited map space. At this time, sections pre-divided by ranking may be allocated according to ranking information for each interest within the community. Alternatively, when the map layout including sections and passages are all changed, the size or location of each section may be determined based on information on the ratio of users by interest. For example, if interests in a specific community consist of a, b, and c, and 60% of users have interest in a, 30% have interest in b, and 10% have interest in c, Sections of interest a, b, and c can be assigned based on the entire map screen or with a size of 6:3:1 of the total area allocated to the section on which the interest is projected within the map screen. Alternatively, as the ratio of users by interest increases, the allocated section can be placed in the very center of the map screen, and as the ratio of users by interest falls, it can be placed in a smaller size at the outskirts. Alternatively, as the ratio of users by interest increases, sections may be allocated and placed in spaces where multiple passages are connected, based on passages.

However, it is not limited to this. The interest section allocation unit 361 may rearrange sections according to the interests of each user. For example, even if among interests a, b, and c, the proportion of users with interest a is high, for users who are only interested in c, interest c rather than a is assigned to the largest section or center of the map screen. It can be displayed.

The section management unit 363 is configured to manage section information within the map screen. For example, the section management unit 363 is configured to manage sections allocated according to interests for each community. Even if it is the same area of interest, it may be managed in a different size or location depending on the community. If the existing section information needs to be changed according to a change in interest information in the section update unit 365, which will be described later, the section management unit 363 ) is configured to expose section information prepared in advance according to the assigned section information.

For example, the section management unit 363 is configured to manage designs exposed on the map screen in advance by size. For example, the section management unit 363 manages information on the same interest in various versions based on the size or location of each section, and displays the corresponding version of interest information on the map screen according to changes in interest information for each community. You can do it. Meanwhile, the interest information may be managed in the central server 300 or in an external server (not shown). For example, in the case of a specific interest, advertisements can be displayed for related products, and related sections can be managed directly by the ad server or server operated by the advertiser. For example, in a section assigned to interest in bicycles, interest information may be displayed by displaying products of a specific bicycle brand.

Meanwhile, the section management unit 363 is configured to manage not only the interest information displayed on the map screen, but also the visit screen on which the interest information displayed on the user screen is projected when the user visits the interest section. In other words, unlike the limited map screen, the visit screen can be commonly managed regardless of the community. For example, depending on interest information, information such as related product introductions and news may be provided, and a chat window may be configured to enable communication between users. There are no size or space restrictions on the landing screen.

The section update unit 365 is configured to update information for each section according to changes in interest information for each community. That is, when the list reconstruction unit 355 is configured to reconstruct the interest list for each community according to a change in the interest information of the user who is a member of each community, it is configured to update the information for each section in the map screen according to the reconstructed interest information.

For example, when a change occurs, such as adding or deleting a user's interests in the community, the assigned section can be changed or a new section can be created or deleted according to the corresponding change in the community's interest list. . In addition, as the user's interest information changes, the interests of each section can be managed by ranking according to the ranking information of interests by community. That is, if there is a change in the ranking information, interests for each section can be reallocated based on the change in ranking.

The output unit 370 may output, to the user terminal 100, information for each section in which interests for each community are projected onto a map screen for each community that matches the group to which each user belongs. In addition, the output unit 370 is configured to output, when a specific user visits a section in which a specific interest is projected, interest-specific content managed for each section on the premise of accessing the metaverse environment.

According to an embodiment of the present invention, by grouping users with the same tendency and providing a plurality of metaverse-based communities in which users are grouped by tendency, users with the same tendency can easily share common interests, and , Although they are not currently interested, the interests of other users with the same tendency are managed as potential interests, so that users can not only have one interest, but also a number of common interests including current and potential interests, a metaverse-based community. Information can be shared and communicated within the space.

According to an embodiment of the present invention, various personal data such as each user's age, taste, personality, and interests are collected, users with high similarity are grouped, and information registered as interests by users in the group is collected. It is possible to provide a metaverse space optimized for groups grouped on the basis of common interests on a map screen where avatars can move or be active so that interests are reflected in the metaverse space according to the inflow or outflow of users with the same tendency for each community. By creating at least one section accordingly and managing the size of the section according to the ratio of users sharing the interest, it is possible to support users to effectively access the space where their interests are projected in the metaverse space.

Hereinafter, with reference to FIG. 6, a method for providing a community according to an embodiment of the present invention will be described. Figure 6 is a flowchart illustrating a method for providing a community according to an embodiment of the present invention.

Referring to FIGS. 1 to 6, the community providing method according to an embodiment of the present invention includes analyzing user tendencies based on collected information for each user (S100), and selecting users based on the analyzed user tendencies. A grouping step (S200), a step of managing a list of interests for each community based on interest information related to each community matched to each group (S300), and allocating sections for each interest on the community map screen based on the list of interests for each community. A step S400 may be included, and a step S500 of reorganizing sections within the map screen according to changes in the list of interests may be further included.

In the step (S100) of analyzing the user's tendency based on the collected information for each user, user information is collected from the user terminal 100.

The user information is provided through the user terminal 100 to general customer information such as name, ID, address, etc., and gender, age, physical condition, income, SNS activity, religion, etc. corresponding to matching information for user targeting. It includes educational background information, etc., and includes various personal data such as personality, taste, aptitude, method, expertise, preference, spending power, activity, and area of interest related to the user's community tendency for user grouping. Includes.

The propensity information may be provided by user input for each item, or may be extracted from other information. For example, the user's area of interest can be extracted by collecting online interest search history, interest information issues, and shopping history provided from the user terminal 100 or the information provision server for each user. This information is personal information, and consent to provision of information may be provided. In addition, it is configured to collect visit history for each section assigned by interest within the community for each user managed by the central server 300. The visit history information is managed as user information.

Accordingly, based on the user information, preference information for each user can be analyzed through a combination of at least one item. For example, the user's tendency information can be analyzed by assigning weight to each information collected through multiple routes. In the case of the weight, weight may be given to information directly input by the user, or if the same information is extracted through multiple routes, weight may be given to the information. The propensity information is an indicator for analyzing the propensity of each user, and is analyzed into items such as personality, taste, aptitude, method, expertise, preference, spending power, activity, and area of interest related to the performance propensity of each user, which is the personalized data described above. It can be. The item may be subdivided into sub-items. For example, spending power can be subdivided according to section, and in the case of interest area items, it can be subdivided into major categories, medium categories, and small categories to which the interests belong. Additionally, each item may be subdivided into specific items or sections. However, it is not limited to this.

Additionally, the information for each item may be expressed as specific information or numerical information. For example, when a user searches for a specific community in the search history, or based on the information entered when registering as a member of the service, user preference information can be extracted based on representative tendencies or interest information in the community. You can. Of course, the information may be represented as at least one piece of information according to priority. For example, in the field of exercise, it can be analyzed whether the user prefers to directly exercise or only watch sports broadcasts. Additionally, specific information among each item of information may be converted into numerical information with predetermined labeling and recorded. For example, in preference items, numerical information pre-specified for each field, for example, '1' may be assigned to exercise, and a numerical value of 'n' may be assigned to music. However, it is not limited to this. In addition, additional quantification can be done according to the section for each detailed item, and the tendency for each subdivided item can be derived through the listed numbers.

Meanwhile, in order to obtain user information, the questionnaire method can be applied, and necessary data can be extracted from answers in subjective form as well as multiple choice form. Accordingly, in order to analyze the user's tendency information based on the input information, text analysis technology based on artificial intelligence can be applied. For example, when learning a model and model for classifying the described text into specific information for each item, the database described above can be used as a training dataset. Accordingly, one embodiment of the present invention is an NLP preprocessor that receives data text from a database as input and analyzes it in units of morphemes to extract only meaningful information in motion classification, and all morphemes extracted from one sentence are classified according to predetermined standards. It may further include a training data generator that labels with information, and a CNN or RNN model that learns the generated training data.

In the step of grouping users based on the analyzed user tendencies (S200), users are classified into similar groups according to the user tendency information that is a combination of the information for each item.

In other words, users can be classified into similar groups in the user grouping unit described later according to the user's tendency information that is a combination of the information for each item. For example, group classification may be based on a combination of 2-dimensional, 3-dimensional, or N-dimensional items depending on the number of items. For example, a classification combining preference item information and interest area item information may be applied. Meanwhile, an item may have sub-items depending on the information. Of course, each sub-item can also be subdivided into additional sub-items.

Meanwhile, for tendency analysis, cohort analysis can be performed based on the user information. For example, the tendencies of each user can be analyzed in a cohort based on the interest information performed by each user, and the tendencies of each user can be analyzed through a combination of at least one cohort item. In other words, each item for each propensity can be treated as a cohort item.

For example, user information can be classified according to a cohort classification combining the above cohort items. For example, the cohort classification may represent information that combines two-dimensional, three-dimensional, or N-dimensional cohort items depending on the number of cohort items. For example, cohort classification combining age information and preference item information may be applied. Meanwhile, a cohort item may have sub-cohort items depending on the information. For example, the age cohort item may consist of Cohort 1 for users in their teens, Cohort 2 for users in their 20s, Cohort 3 for users in their 30s, etc. Of course, each Cohort 1, Cohort 2, and Cohort 3 can be subdivided into subcohorts such as Cohort 1a, Cohort 1b, and Cohort 1c according to detailed classification criteria (e.g., early, mid, and late) within each cohort. You can. Likewise, Cohort 1a can also be subdivided again. In other words, when analyzing user tendencies according to N cohort items, each user can be classified as having tendencies classified by N cohorts. Meanwhile, priority may also be given to cohort items. That is, users can be grouped based on the tendency information for each item ranging from first priority to a predetermined priority according to the priority of each user. For example, users can be output as a scatterplot based on some cohort items according to priority among multiple cohort items. In other words, users can be placed on the graph according to the priority of the cohort item, and when it is necessary to divide users into groups, users can be further classified according to the next-priority cohort item.

In this step, users are grouped into groups composed of users with a certain degree of similarity based on the analyzed tendency information for each user. For example, based on information for each item, users who share at least one piece of information can be grouped into the same group. Additionally, when a large number of users are concentrated in one group, the information for each item shared in the grouping can be subdivided and regrouped. In other words, if the user group is biased to one side and it is difficult to classify it into a certain number of groups, the biased group is set as the analysis range and the users in that range are regrouped again. In addition, if additional regrouping is required for the regrouped user group, the regrouping can be done in the same manner.

Meanwhile, in user regrouping, various methods can be applied. For example, user groups may be initially grouped according to higher-level items, and if regrouping is required, the user groups can be regrouped by subdividing them into lower-level items. That is, in the case of users grouped with the same or similar tendencies based on the priority of the preference item, etc., they may be regrouped according to the priority of subordinate items of the preference item. Alternatively, in a user group represented by scattered points with n-dimensional coordinates according to the number of items or information shared by tendency information by the coordinate method, if regrouping is required, items or information are added to create an n + 1-dimensional By representing scattered points with coordinates, users can be regrouped based on the distance between the scattered points. For example, if regrouping is necessary for a user group grouped based on the user's personality and preferences, the group can be regrouped by adding interest items that are the next priority after the personality and preferences. Of course, if regrouping is required even after this, regrouping can be performed by adding items or information of the next rank and expressing them in n + 2-dimensional coordinates.

At this time, the group concept may be a method of recommending a group similar to the user's tendency information or similar to the user's tendency by calculating similarity according to the group's tendency based on the group's profile. At this time, if the group's profile information is clear and has various propensity values, it may be implemented as a content-based group recommendation method rather than collaborative filtering. In other words, a group is not a fixed concept, and since it is grouped based on the tendency information of a plurality of users, it should be understood as a concept that can vary depending on the grouped users. This user group method is intended to recommend interests within the group to which each user belongs, taking into account the individual's tendencies and the tendencies of other users with the same tendencies, in order to ensure satisfaction for each user. Accordingly, information reflecting trends can be managed according to the influx of new users organized within the same group, and users belonging to the existing group can also be recommended more diverse interests as new users flow in.

First, in grouping, propensity information and the Euclidean calculation method can be used to calculate the average and standard deviation of the similarity of the propensity information for each user. At this time, the average and standard deviation can be used to calculate the central value for each individual's propensity, and multidimensional (propensity information) values can be calculated to calculate distance based on the central value for each propensity. Additionally, distance-based similarity between groups can be calculated. If the distance between groups is more than a certain level, they can be grouped into different groups, and if the distance between groups is less than a certain level, they can be grouped into the same group. Of course, various methods other than those described above may be used. At this time, the meaning of the above-mentioned group may mean a group of individuals with similar demographic tendencies.

In the step (S300) of managing a community-specific interest list based on interest information that is correlated to each community matched to each group, big data of the analyzed user-specific tendency information of a large number of users and the individual interest information of each user are analyzed. Thus, it is configured to extract the correlation for each information about the tendency information of a large number of users and each interest information. In other words, it is configured to extract the correlation for each information from item information and interest information from the tendency information for each user. Accordingly, detailed information that influenced actual interests can be extracted according to each user's tendencies.

Meanwhile, the extraction unit for this embodiment is not limited to information for each item of the same nature in extracting the correlation between user-specific tendency information and interest information. For example, if there is a priority for a specific item of user-specific tendency information, it is limited to the specific item and common information is not extracted, and if a specific interest is common from multiple interest information, the interest related to the interest field item is not extracted. Common information can be calculated.

In other words, extract the correlation of common information for each detailed item focusing on items with high correlation with information about the community, including primarily related information, but also extract information on different items where the presence or absence of correlation is secondarily suspected. Try to extract correlation. In other words, the correlation between the tendency and at least one piece of unspecified information is extracted from the interest information of multiple users with the same tendency. The correlation may be determined by the frequency of each information. Among the interest information of multiple users with a specific tendency, information with a high frequency is extracted with high correlation, and in the case of a community with the combined information, if it contains a certain number, it is matched to the specific tendency. As a result, one community may be matched with several groups based on similarity.

Meanwhile, a high correlation or correlation of a specific item means that it is higher than that of other items or has a correlation or correlation above a certain value, which is a predetermined standard.

When an item with a high correlation with specific propensity information and specific information about the item are extracted from the extraction unit 330, a relationship diagram between the item, specific information about the item, and the specific propensity can be generated. In other words, when users are grouped by propensity by the user grouping unit described later based on propensity information, nodes are formed for each grouped propensity, and a relationship diagram in which each interest information is connected through a network as a node can be created. . For example, it may be possible to connect to a network only when nodes have a certain level of quantified correlation.

Meanwhile, the extraction unit 330 learns the correlation between each segmented item or information based on the tendency information and interest information of an unspecified number of people based on artificial intelligence, and creates a database of the correlation between the items and information, Each time data is updated, the correlation may be configured to update. Accordingly, based on an artificial intelligence-based correlation analysis model, a community containing interest information with a high correlation to the analyzed user tendency is recommended, or when a new community is registered, interest information is analyzed to determine a specific tendency. It can be matched to a user group with .

When a user is grouped, it is configured to match a matching community for each classified group based on the user's tendency information. In addition, it is configured to manage the interests of users belonging to each community and reorganize the interest list according to the inflow or outflow of users within the community. Users may be grouped according to each user's tendency, and communities may be matched for each group, or multiple groups may be matched into one community. At this time, the groups that are matched to one community are based on the premise that the user's tendency information or interest information belonging to each group has a degree of similarity above a certain level.

In this step, the interest list of the community to which the highly correlated interest information extracted by the extraction unit 320 is matched is managed for each community to which the group classified based on the propensity information is matched. That is, in a group classified based on user-specific tendency information, highly correlated information is extracted as community interest information from the user information of users in the group, and a highly consistent interest list containing highly correlated information for each community. It is configured to manage. In other words, the interest list is matched for each community. For example, in the case of community A, the first interest list, and in the case of community B, the second interest list may be matched and managed.

In addition, a list of at least one interest can be generated and output according to the ranking of each community depending on whether it contains a lot of highly correlated information. For example, interests containing highly correlated information can be output for each community, and the rankings can be sorted according to the number of matches of highly correlated information. For example, according to the ranking in which multiple pieces of highly correlated information are matched, they may be arranged as 1. a interest, 2. b interest, 3. d interest, etc. In other words, a user belonging to a specific community can sign up for the service and receive a recommended list of interests at the time of access.

Additionally, it is configured to manage interest information for each community based on interest information for each user. For example, if user ‘A’ is interested in a, b, and c within a specific community, and user ‘B’ is interested in a, d, and f, the interest list management unit 353 will: As interests of a specific community, a, b, c, d, and f can be managed as a list, and a can be managed by treating it as a priority as a common interest. At this time, d and f can be managed as potential interests for 'A', and a and b can be managed as potential interests for 'B'. Meanwhile, as interest information, it is managed as information according to major, medium, and small categories for each item of interest, and is configured to manage information according to community information changes, updates, and promotions in real time. For example, if a new subscriber, 'Sick', is grouped with the same tendency and matched to the same community, if 'Sick''s interests are a and k, k can be registered and managed as a new interest in the community. .

In addition, a list of similar interests can be managed according to the similarity of each community based on tendency information. For example, according to the similarity based on the tendency information of each user belonging to each community, each matching interest list can be managed as a similar interest list. In the above example, when the preference information of users belonging to Community A and Community B have a high degree of similarity for each item, the first interest list and the second interest list may be managed as similar interest lists. The similarity for each group may be determined depending on whether the propensity information for each item matches. In addition, in the case of a user group represented by scattered points with n-dimensional coordinates according to the number of items or information shared for each tendency information by the coordinate method, the degree of similarity can be determined according to the proximity of the group shown on the coordinates.

Accordingly, it is possible to create a list of interests that primarily matches the community to which the user belongs, and if there is no object of interest tailored to the user in the list of interests, interest lists with high similarity are sequentially included, allowing the user to participate. We can support you to do so.

In the step (S400) of allocating a section for each interest on the community map screen based on the list of interests for each community, a section in which the user's interests are projected is assigned to the map screen for each community. The map screen is an area where users of each community can move or engage in activities in the metaverse space, assuming that it is a limited map space.

In other words, it is configured to allocate sections by interest to certain areas so that users can visit them in a limited map space. At this time, sections pre-divided by ranking may be allocated according to ranking information for each interest within the community. Alternatively, when the map layout including sections and passages are all changed, the size or location of each section may be determined based on information on the ratio of users by interest. For example, if interests in a specific community consist of a, b, and c, and 60% of users have interest in a, 30% have interest in b, and 10% have interest in c, Sections of interest a, b, and c can be assigned based on the entire map screen or with a size of 6:3:1 of the total area allocated to the section on which the interest is projected within the map screen. Alternatively, as the ratio of users by interest increases, the allocated section can be placed in the very center of the map screen, and as the ratio of users by interest falls, it can be placed in a smaller size at the outskirts. Alternatively, as the ratio of users by interest increases, sections may be allocated and placed in spaces where multiple passages are connected, based on passages.

However, it is not limited to this. Sections can be rearranged according to user interests. For example, even if among interests a, b, and c, the proportion of users with interest a is high, for users who are only interested in c, interest c rather than a is assigned to the largest section or center of the map screen. It can be displayed.

Additionally, this step is configured to manage section information within the map screen. For example, each community is configured to manage sections assigned according to interests. Even if it is the same area of interest, it may be managed in a different size or location depending on the community. If the existing section information needs to be changed due to a change in the interest information described later, section information prepared in advance is exposed according to the assigned section information. It is configured to do so.

For example, the design exposed on the map screen is configured to manage in advance by size. For example, information on the same interest can be managed in various versions based on the size or location of each section, and corresponding versions of interest information can be displayed on the map screen according to changes in interest information for each community. Meanwhile, the interest information may be managed in the central server 300 or in an external server (not shown). For example, in the case of a specific interest, advertisements can be displayed for related products, and related sections can be managed directly by the ad server or server operated by the advertiser. For example, in a section assigned to interest in bicycles, interest information may be displayed by displaying products of a specific bicycle brand.

Meanwhile, it is configured to manage not only the interest information displayed on the map screen, but also the visit screen on which the interest information displayed on the user screen is projected when the user visits the interest section. In other words, unlike the limited map screen, the visit screen can be commonly managed regardless of the community. For example, depending on interest information, information such as related product introductions and news may be provided, and a chat window may be configured to enable communication between users.

In the step of reorganizing sections within the map screen according to changes in the interest list (S500), information for each section is updated according to changes in interest information for each community. That is, when the interest list for each community is configured to be reorganized according to a change in the interest information of a user who is a member of each community, the information for each section in the map screen is updated according to the reconstructed interest information.

For example, when a change occurs, such as adding or deleting a user's interests in the community, the assigned section can be changed or a new section can be created or deleted according to the corresponding change in the community's interest list. . In addition, as the user's interest information changes, the interests of each section can be managed by ranking according to the ranking information of interests by community. That is, if there is a change in the ranking information, interests for each section can be reallocated based on the change in ranking.

The community provision method according to an embodiment of the present invention described above can be executed by an application installed by default on the terminal (this may include programs included in the platform or operating system, etc. installed by default on the terminal), and the user It may also be executed by an application (i.e. program) installed directly on the master terminal through an application providing server such as an application store server, a web server related to the application or the service. In this sense, the community provision method according to an embodiment of the present invention described above is implemented as an application (i.e., a program) installed by default in the terminal or directly installed by the user and stored in a computer-readable recording medium such as the terminal. can be recorded

The description of the present invention described above is for illustrative purposes, and those skilled in the art will understand that the present invention can be easily modified into other specific forms without changing the technical idea or essential features of the present invention. will be. Therefore, the embodiments described above should be understood in all respects as illustrative and not restrictive. For example, each component described as unitary may be implemented in a distributed manner, and similarly, components described as distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the claims described below rather than the detailed description above, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention. do.

The above-described present invention can be implemented as computer-readable code (or application or software) on a program-recorded medium. The community provision method described above can be realized by code stored in memory, etc.

Computer-readable media includes all types of recording devices that store data that can be read by a computer system. Examples of computer-readable media include HDD (Hard Disk Drive), SSD (Solid State Disk), SDD (Silicon Disk Drive), ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc. It also includes those implemented in the form of carrier waves (e.g., transmission via the Internet). Additionally, the computer may include a processor. Accordingly, the above detailed description should not be construed as restrictive in all respects and should be considered illustrative. The scope of the present invention should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of the present invention are included in the scope of the present invention.

100: user terminal
300: central server
310: collection department
320: Analysis department
330: extraction unit
340: User grouping unit
350: Community Management
360: Map screen management unit
370: output unit

Claims (10)

As a method of providing community in the metaverse environment,
Analyzing user-specific tendency information according to predetermined items from the collected user information;
grouping users based on user preference information for each item of at least one of the predetermined items;
Managing an interest list for each community based on interest information of users in the community matching at least one group;
Based on the interest list for each community, allocating a section for each interest on a map screen for each community in the metaverse environment; and
Comprising a step of reconstructing an interest-specific section of the community-specific map screen according to a change in the community-specific interest list according to a change in the interest information of the user for each community,
The step of managing the interest list for each community is,
When matching interests for each user in the at least one group, interests registered by other users in the same group or extracted for the other users are matched with potential interests,
A method for providing a community, wherein when a new subscriber is added to a specific group, interests of the new subscriber that are not in the interest list of the community matched to the specific group are registered as new interests. According to claim 1,
By analyzing the analyzed user-specific tendency information of a large number of users in the community and the big data of the user-specific interest information, at least one interest information that is correlated to each community is extracted, and the community-specific interest list includes the correlation. A community providing method characterized by being based on at least one interest information. According to claim 1,
A community providing method, characterized in that the list of interests for each community changes according to user interest information flowing in and out of each community. According to claim 1,
A method for providing a community, wherein the list of interests for each community changes according to changes in interest information of users for each community. According to claim 1,
A method for providing a community, wherein the interest list for each community includes ranking information for each interest based on the ratio information of users for each interest. According to clause 5,
A method for providing a community, wherein the size or location of each interest section is determined based on the interest ranking information. According to clause 5,
A method for providing a community, characterized in that the interest-specific sections are allocated to sections pre-divided by ranking based on the interest-specific ranking information. According to claim 1,
The predetermined item includes at least one of personality, taste, aptitude, method, expertise, preference, spending power, activity, and interest area related to the user's inclination, and the user's inclination is analyzed by a combination of the information for each item. A community provision method characterized by being. According to claim 1,
The step of grouping the users is a method of providing a community, wherein the users are grouped based on tendency information for each item ranging from first priority to a predetermined priority according to the priority of each user. According to clause 9,
The step of grouping users includes, when a large number of users are concentrated in a specific group, adding tendency information of items of the next highest priority in the predetermined ranking to regroup users included in the specific group. method.