A method for generating an outlier data, and an electronic device and system for performing such a method

The virtual environment platform addresses inefficiencies in data generation and trading by allowing data providers to generate data meeting user needs and ensuring secure transactions, enhancing AI model training and transaction stability.

KR103044667B1Active Publication Date: 2026-11-06PEBLUS CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
KR1020230040249
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2023-03-28
Publication Date
2026-11-06
Estimated Expiration
2042-10-18

AI Technical Summary

Technical Problem

Current data generation platforms rely on crowdsourcing, leading to inefficiencies and instability in data transactions due to centralized ownership records, making it difficult to match user requirements and ensure secure data trading.

Method used

A virtual environment platform that allows data providers to generate data through a processor-driven method, ensuring data sets meet user requirements, and facilitates secure data transactions using blockchain technology to manage ownership and rewards.

Benefits of technology

Enables data generation that meets user needs and secure data trading, optimizing AI model training and improving transaction stability and efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure R1020230040249_ABST
    Figure R1020230040249_ABST
Patent Text Reader

Abstract

According to one embodiment of the present disclosure, an electronic device may be provided comprising a memory and at least one processor electronically connected to the memory and operating based on at least one instruction stored in the memory, wherein the at least one processor obtains a first generation condition for generating data based on a reference data set previously stored in the memory—wherein the first generation condition is set based on similarity with the reference data set—obtains a first synthetic data generated from a first user terminal, identifies a first association between the first synthetic data and the reference data set based on similarity between the first synthetic data and the reference data set, and determines the first synthetic data as outlier data based on a judgment that the identified first association corresponds to the first generation condition.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present disclosure relates to an electronic device (or computing device) that provides a comprehensive data platform equipped with various functions related to data in a virtual environment. Background Technology

[0002] As artificial intelligence technology advances, research on AI models is actively underway. However, along with the structure of AI models, training data is crucial for model learning. Accordingly, research on synthetic data (or virtual data) generated by specific methods, rather than actual data, is actively being conducted to acquire high-quality, large-scale training data.

[0003] With data collection becoming increasingly important, there is a growing trend of data generation platforms emerging to collect large volumes of data. However, current data generation platforms operate on a crowdsourcing basis, requiring each data provider to secure their own data; consequently, they fail to provide an environment where data can be generated directly.

[0004] Furthermore, as the value of data as an asset increases, data trading is becoming more active. However, since data is an intangible asset that must be traded through networks, stability is the most critical factor in transactions. Yet, when data trading takes place through a centralized network, records regarding data ownership are concentrated in the center, making it difficult to achieve the aforementioned transaction stability. Additionally, inefficiencies arise in the data trading process, such as requiring data users to directly search for data with desired attributes in order to purchase the data they want.

[0005] The present disclosure provides a virtual environment platform for solving various problems arising in the data generation environment and transaction environment described above. The problem to be solved

[0006] The problem to be solved in this disclosure is to provide an environment capable of generating data that matches the requirements of users.

[0007] Another problem to be solved in this disclosure is to provide an environment where data can be traded securely.

[0008] Meanwhile, the problems to be solved in this disclosure are not limited to those described above, and problems not mentioned will be clearly understood by those skilled in the art to which the invention included in this disclosure belongs from this specification and the attached drawings. means of solving the problem

[0009] According to one embodiment of the present disclosure, a method for providing a virtual environment in which at least one data provider terminal can generate data may be provided, comprising: a step of obtaining a request for the generation of a data set by at least one processor of an electronic device; a step of checking a reference data set previously stored in memory based on the obtained request; a step of obtaining first request information including a first generation condition set based on a relationship with the reference data set, and a step of providing a first virtual environment for generating data based on the first request information; a step of checking whether a first synthetic data set generated by at least one user terminal accessing the first virtual environment satisfies the first generation condition; and a step of blocking user access to the first virtual environment based on a determination result that the first synthetic data set satisfies the first generation condition.

[0010] Additionally, according to one embodiment of the present disclosure, a method for performing data transactions in a virtual environment may be provided, comprising: a step of obtaining a first sub-data set from a first user terminal and obtaining a second sub-data set from a second user terminal by at least one processor of an electronic device; a step of obtaining a first data set including the first sub-data set and the second sub-data set; a step of obtaining contribution information reflecting the contribution of the first user terminal and the contribution of the supplier terminal of the second user terminal based on the first sub-data set and the second sub-data set; a step of obtaining a first block data based on the first data set and the contribution information; a step of obtaining a second block data based on a transaction agreement corresponding to the first data set; and a step of distributing a first reward generated by the transaction agreement to the first user terminal and the second user terminal based on the contribution information.

[0011] Additionally, according to one embodiment of the present disclosure, a method of operation of an electronic device for optimizing an artificial intelligence model may be provided, comprising: a step of constructing a virtual environment by at least one processor of the electronic device; a step of providing at least one data generation tool to at least one user terminal accessing the constructed virtual environment; a step of obtaining a first synthetic data set using the at least one generation tool; a step of learning a reference model stored in the memory of the electronic device based on the first synthetic data set; a step of inputting a first evaluation data set to the learned reference model; and a step of providing feedback to the at least one user terminal when a result output from the reference model based on the first evaluation data set does not satisfy a predetermined criterion, wherein the feedback is configured to request the generation of data associated with the first evaluation data set.

[0012] The means for solving the problem of the present invention are not limited to the means described above, and means not mentioned will be clearly understood by those skilled in the art to which the present invention pertains from this specification and the accompanying drawings. Effects of the invention

[0013] According to embodiments of the present disclosure, an electronic device may be provided that provides an environment capable of generating data that matches the requirements of a user.

[0014] In addition, according to embodiments of the present disclosure, an electronic device that provides an environment for safely trading data may be provided.

[0015] The effects of the present invention are not limited to those described above, and unmentioned effects will be clearly understood by those skilled in the art from this specification and the accompanying drawings. Brief explanation of the drawing

[0016] FIG. 1 is a drawing illustrating a system providing a data platform according to various embodiments. FIG. 2 is a diagram illustrating the hardware configuration of an electronic device (e.g., a platform providing device or a server and / or administrator terminal included in the platform providing device) according to various embodiments. FIG. 3 is a diagram illustrating various environments provided in a data platform environment provided by an electronic device according to various embodiments. FIG. 4 is a diagram illustrating the hardware configurations of an electronic device (e.g., a user terminal such as a mobile phone, PC, etc.) according to various embodiments. FIG. 5 is a diagram illustrating examples in which a data platform environment is implemented through an electronic device according to various embodiments. FIG. 6 is a diagram illustrating operations performed by an electronic device in a data platform environment according to various embodiments. FIG. 7 is a diagram illustrating information that can be generated or received by an electronic device and stored in memory according to various embodiments. FIG. 8 is a flowchart illustrating a method in which at least one electronic device generates a data set according to various embodiments. FIG. 9 is a diagram illustrating, according to various embodiments, an exemplary method in which at least one electronic device generates a data set. FIG. 10 is a diagram illustrating the operation of an electronic device updating information of a project for data generation according to various embodiments. FIG. 11 is a diagram illustrating a method of performing a data generation operation by providing a data image to an electronic device according to various embodiments. FIG. 12 is a drawing illustrating a user interface in which an electronic device generates data using a data generation tool according to various embodiments. FIG. 13 is a flowchart illustrating a method of visualizing and providing data generated by an electronic device according to various embodiments. FIG. 14 is a drawing illustrating a user interface that visualizes and provides data generated by an electronic device according to various embodiments. FIG. 15 is a flowchart illustrating information that an electronic device may additionally provide in relation to data generation according to various embodiments. FIG. 16 is a drawing illustrating an embodiment of a user interface in which an electronic device provides additional information regarding data generation according to various embodiments. FIG. 17 is a drawing illustrating another embodiment of a user interface in which an electronic device provides additional information regarding data generation according to various embodiments. FIG. 18 is a diagram illustrating the function of an electronic device to provide a data trading environment according to various embodiments. FIG. 19 is a diagram illustrating how an electronic device provides a data trading environment according to various embodiments. FIG. 20 is a flowchart illustrating a method for an electronic device to trade a synthetic data set obtained from a supplier terminal using a blockchain network according to various embodiments. FIG. 21 is a diagram illustrating information stored in a blockchain network as an electronic device trades a synthetic data set obtained from a supplier terminal according to various embodiments. FIG. 22 is a flowchart illustrating a method in which an electronic device provides a smart contract-based data trading environment according to various embodiments. FIG. 23 is a diagram illustrating a method for generating block data accompanying a smart contract-based data transaction using an electronic device, according to various embodiments. FIG. 24 is a diagram illustrating a general method for an electronic device to acquire contribution information according to various embodiments. FIG. 25 is a diagram illustrating a method for an electronic device to acquire contribution information according to quality information according to various embodiments. FIG. 26 is a diagram illustrating a method for an electronic device to generate quality information for a generated data set according to various embodiments. FIG. 27 is a flowchart illustrating a method for an electronic device to acquire platform contribution information according to various embodiments. FIG. 28 is a diagram illustrating a method for improving data of an electronic device providing a platform according to various embodiments. FIG. 29 is a diagram illustrating a method for obtaining platform contribution information as an electronic device providing a platform improves data according to various embodiments. FIG. 30 is a diagram illustrating a method in which an electronic device provides feedback on data generation using a reference model according to various embodiments. FIG. 31 is a diagram illustrating how an electronic device learns and demonstrates a waste plastic classification model according to various embodiments. FIG. 32 is a flowchart illustrating a data construction method for an electronic device to learn a model for identifying outlier data according to various embodiments. FIG. 33 is a diagram illustrating a method for an electronic device to demonstrate a model for determining outlier data according to various embodiments. Specific details for implementing the invention

[0017] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In describing the embodiments, technical details that are well known in the art to which the present disclosure belongs and are not directly related to the present disclosure will be omitted. This is intended to convey the essence of the present disclosure more clearly without obscuring it by omitting unnecessary explanations.

[0018] The embodiments described in this specification are intended to clearly explain the concept of the invention to those skilled in the art to which the invention pertains, and therefore the invention is not limited to the embodiments described in this specification, and the scope of the invention should be interpreted to include modifications or variations that do not deviate from the concept of the invention.

[0019] The terms used in this specification have been selected to be as widely used as possible, taking into account their functions in the present invention; however, these terms may vary depending on the intent of those skilled in the art to which the present invention pertains, case law, or the emergence of new technologies. However, if a specific term is defined and used with an arbitrary meaning, the meaning of that term will be described separately. Accordingly, the terms used in this specification should be interpreted based on their actual meaning and the content throughout this specification, rather than merely their names.

[0020] The drawings attached to this specification are intended to facilitate the explanation of the present invention. The shapes depicted in the drawings may be exaggerated as necessary to aid in understanding the present invention, and therefore the present invention is not limited by the drawings.

[0021] In cases where it is determined that a detailed description of known components or functions related to the present invention in this specification may obscure the essence of the invention, such detailed description shall be omitted as necessary. Furthermore, numbers used in the description of this specification (e.g., First, Second, etc.) are merely identification symbols to distinguish one component from another.

[0022] Furthermore, the suffixes "part" and "substance" used for components in the following description are assigned or used interchangeably solely for the ease of drafting the specification, and do not inherently possess distinct meanings or roles.

[0023] That is, the embodiments of the present disclosure are provided to make the present disclosure complete and to inform those skilled in the art of the scope of the present disclosure, and the invention of the present disclosure is defined only by the scope of the claims. Throughout the specification, the same reference numerals refer to the same components.

[0024] Terms such as “first” and / or “second” may be used to describe various components, but said components shall not be limited by said terms. For the sole purpose of distinguishing one component from another, for example, without departing from the scope of rights according to the concept of the present disclosure, the first component may be named the second component, and similarly, the second component may be named the first component.

[0025] When it is stated that one component is "connected" or "connected" to another component, it should be understood that while it may be directly connected or connected to that other component, there may also be other components in between. Conversely, when it is stated that one component is "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between. Other expressions describing the relationship between components, such as "between" and "exactly between," or "adjacent to" and "directly adjacent to," should be interpreted in the same way.

[0026] In the drawings, each block of the process flow diagrams and combinations of the flow diagrams can be executed by computer program instructions. Since these computer program instructions can be loaded into the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing equipment, the instructions executed through the processor of the computer or other programmable data processing equipment create means to perform the functions described in the flow diagram block(s). Since these computer program instructions can also be stored in computer-available or computer-readable memory that can be directed toward the computer or other programmable data processing equipment to implement the function in a specific way, the instructions stored in computer-available or computer-readable memory can also produce a manufactured item containing means of instruction to perform the function described in the flow diagram block(s). Since computer program instructions can be loaded onto a computer or other programmable data processing equipment, instructions that execute a computer or other programmable data processing equipment by performing a series of operation steps on the computer or other programmable data processing equipment to create a process executed by the computer may also provide steps for executing the functions described in the flowchart block(s).

[0027] Additionally, each block may represent a module, segment, or part of code containing one or more executable instructions for executing a specified logical function(s). It should also be noted that in some alternative execution examples, the functions mentioned in the blocks may occur out of order. For instance, two blocks described in succession may actually be executed substantially simultaneously, or the blocks may be executed in reverse order according to their corresponding functions.

[0028] As used in this disclosure, the term “unit” refers to a software or hardware component such as a Field Programmable Gate Array (FPGA) or an Application Specific Integrated Circuit (ASIC). A “unit” performs specific roles but is not limited to software or hardware. A “unit” may be configured to reside in an addressable storage medium or may be configured to run one or more processors. Accordingly, according to some embodiments, a “unit” includes components such as software components, object-oriented software components, class components, and task components, as well as processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. The functions provided within the components and “units” may be combined into a smaller number of components and “units” or further separated into additional components and “units.” In addition, the components and 'parts' may be implemented to utilize one or more CPUs within the device or secure multimedia card. Furthermore, according to various embodiments of the present disclosure, the 'parts' may include one or more processors.

[0029] The operating principles of the present disclosure will be described in detail below with reference to the attached drawings. In describing the present disclosure below, specific descriptions of related known functions or configurations will be omitted if it is determined that such detailed descriptions would unnecessarily obscure the essence of the present disclosure. Furthermore, the terms described below are defined in consideration of their functions in the present disclosure, and these may vary depending on the intentions or practices of the user or operator. Therefore, their definitions should be based on the content throughout this specification.

[0030] According to one embodiment of the present disclosure, a method for providing a virtual environment in which at least one data provider terminal can generate data may be provided, comprising: a step of obtaining a request for the generation of a data set by at least one processor of an electronic device; a step of checking a reference data set previously stored in memory based on the obtained request; a step of obtaining first request information including a first generation condition set based on a relationship with the reference data set, and a step of providing a first virtual environment for generating data based on the first request information; a step of checking whether a first synthetic data set generated by at least one user terminal accessing the first virtual environment satisfies the first generation condition; and a step of blocking user access to the first virtual environment based on a determination result that the first synthetic data set satisfies the first generation condition.

[0031] Additionally, the above request may include at least one of the attributes of the data set—the attributes of the data set include the domain of the data and the modality of the data—the quality of the data set or sample data.

[0032] Additionally, the step of verifying the reference data set may include the step of searching for a reference data set corresponding to the request among a plurality of reference data sets stored in advance in the memory.

[0033] In addition, the first generation condition can be set based on similarity with the reference data set.

[0034] In addition, the first request information may include the first generation condition, the reference data set, and the sample data.

[0035] Additionally, after providing the first virtual environment, the method may further include the step of obtaining level information indicating the level of data to be generated based on the first request information, selecting a first data provider corresponding to the level information based on the level information, and providing information about the first virtual environment to a first user terminal corresponding to the selected first data provider.

[0036] Additionally, the step of checking whether the first generation condition is satisfied may include: a step of checking the similarity between the first synthetic data set and the reference data set; and a step of checking whether the similarity corresponds to the first generation condition.

[0038] In addition, the method further includes the step of obtaining a second generation condition set based on a relationship with the reference data set; wherein the first generation condition is set based on a first association relationship with the reference data set, and the second generation condition can be set based on a second association relationship with the reference data set.

[0039] Additionally, the method may further include the step of providing the first generation condition to a second user terminal accessing the first virtual environment and providing the second generation condition to a third user terminal accessing the first virtual environment.

[0040] Additionally, the method may further include the step of requesting additional data generation from at least one user terminal based on the determination that the first synthetic data set does not satisfy the first generation condition.

[0041] Additionally, it may further include the step of obtaining an improved synthetic data set based on a second synthetic data set additionally generated from at least one user terminal.

[0042] Additionally, the step of providing the first virtual environment may include the step of selecting some of the plurality of data generation tools stored in the memory based on the first request information.

[0043] Additionally, the step of blocking user access to the first virtual environment may include the step of storing the first synthetic data set in the memory.

[0044] Additionally, according to one embodiment of the present disclosure, a method for performing data transactions in a virtual environment may be provided, comprising: a step of obtaining a first sub-data set from a first user terminal and obtaining a second sub-data set from a second user terminal by at least one processor of an electronic device; a step of obtaining a first data set including the first sub-data set and the second sub-data set; a step of obtaining contribution information reflecting the contribution of the first user terminal and the contribution of the supplier terminal of the second user terminal based on the first sub-data set and the second sub-data set; a step of obtaining a first block data based on the first data set and the contribution information; a step of obtaining a second block data based on a transaction agreement corresponding to the first data set; and a step of distributing a first reward generated by the transaction agreement to the first user terminal and the second user terminal based on the contribution information.

[0045] Additionally, the first block data may include first transaction information including information related to the creation of the first data set, and the first transaction information may include ownership information indicating the owner of the first data set, share information indicating the share ratio between co-owners of the first data set, and reward information reflecting a second reward generated by the creation of the first data set.

[0046] In addition, the ownership information may include information about the current owner of the first data set and information about past owners.

[0047] In addition, the above share information is obtained based on the above contribution information, and can be obtained by calculating the share ratio of the first user terminal and the share ratio of the second user terminal based on the contribution of the first user terminal and the contribution of the second user terminal confirmed based on the above contribution information.

[0048] Additionally, the above transaction agreement may include an operation of confirming a transaction request received from a third user terminal and an operation of providing an approval intent to the third user terminal in response to the transaction request.

[0049] Additionally, the above transaction agreement may include an operation of confirming a transaction proposal received from a third user terminal and an operation of providing an intention to approve to the third user terminal in response to the transaction request.

[0050] Additionally, the transaction agreement may include the operation of transmitting a transaction proposal for the first data set to a third user terminal and the operation of receiving an intention to approve the transaction proposal from the third user terminal.

[0051] Additionally, the second block data may include second transaction information that includes information related to the transaction of the first data set, and the second transaction information may include owner change information indicating the status of change of owner according to the transaction and reward information that includes the first reward accompanying the transaction.

[0052] Additionally, the step of distributing to the first user terminal and the second user terminal may include the step of distributing the first reward based on the share ratio of the first user terminal and the share ratio of the second user terminal confirmed by the share information.

[0053] In addition, the contribution information may include platform contribution information obtained based on the extent to which at least one tool provided in the virtual environment is used to generate the first data set.

[0054] In addition, the above at least one tool may include a data generation tool used to generate synthetic data and a quality evaluation tool for evaluating the quality of the data.

[0055] Additionally, the method may further include the step of transmitting the first block data and the second block data to a blockchain network.

[0056] Additionally, according to one embodiment of the present disclosure, a method of operation of an electronic device for optimizing an artificial intelligence model may be provided, comprising: a step of constructing a virtual environment by at least one processor of the electronic device; a step of providing at least one data generation tool to at least one user terminal accessing the constructed virtual environment; a step of obtaining a first synthetic data set using the at least one generation tool; a step of learning a reference model stored in the memory of the electronic device based on the first synthetic data set; a step of inputting a first evaluation data set to the learned reference model; and a step of providing feedback to the at least one user terminal when a result output from the reference model based on the first evaluation data set does not satisfy a predetermined criterion, wherein the feedback is configured to request the generation of data associated with the first evaluation data set.

[0057] In addition, the feedback may include a range of generation parameters for generating data associated with the first evaluation data set—the generation parameters are used by the data generation tool to generate synthetic data.

[0058] In addition, the feedback may include a message requesting the generation of data associated with the first evaluation data set.

[0059] In addition, the first evaluation data set may be provided by a first user terminal that is not allowed access to the virtual environment.

[0060] Additionally, it may further include the step of retraining the reference model based on the second synthetic data set generated based on the above feedback.

[0061] Additionally, the method may further include the step of inputting a second evaluation data set into the retrained reference model; and the step of storing the second synthetic data set in memory when the result output from the reference model based on the second evaluation data set satisfies the predetermined criteria.

[0062] In addition, the above reference model may be configured as a model for identifying recyclable plastics.

[0063] Additionally, the evaluation data set includes first plastic data associated with recyclable plastic, and the at least one processor may be configured to request the at least one user terminal to generate data associated with the first plastic data if, according to the result output from the reference model, the first plastic data is not determined to be recyclable plastic.

[0064] Additionally, the evaluation data set includes second plastic data associated with recyclable plastic, and the at least one processor may be configured to provide a visual effect to the second plastic data when the second plastic data is determined to be recyclable plastic according to the result output from the reference model.

[0065] Hereinafter, an electronic device providing a virtual environment platform including various functions related to the data of the present disclosure, a method of operating the electronic device, and a system including the electronic device are described.

[0067] [System]

[0068] FIG. 1 is a drawing illustrating a system providing a data platform according to various embodiments.

[0069] A system providing a data platform can implement a platform environment that allows users to generate data (e.g., synthetic data), trade data, or receive various experiences utilizing data. Additionally, an electronic device providing a data platform (e.g., a server or administrator terminal) can build a data asset storage by converting and storing various forms of data obtained from users utilizing the platform environment into assets (e.g., synthetic data or data related to the generation and trading of data).

[0070] Referring to FIG. 1, a system (100) providing a data platform may include a platform providing device (110) and at least one user terminal (120a, 120b, ?, 120n) utilizing a data platform environment (150) established by the platform providing device. Additionally, the platform providing device (110) and at least one user terminal (120a, 120b, ?, 120n) may transmit and receive data to and from each other when a communication connection is established by a communication network (Net).

[0071] The platform providing device (110) can provide a data platform environment (150). In this case, the data platform environment (150) may be a virtual environment for providing various functions and services related to data, such as generating, trading, or utilizing data.

[0072] A platform providing device (110) may include a server (111) and at least one administrator terminal (113) for managing the server. Specifically, various platform environments provided by the server (111) can be managed through at least one administrator terminal (113), and furthermore, the administrator terminal (113) can perform a specific role in the various platform environments (e.g., a role such as a 'producer' designing a project for data generation). In addition, not limited thereto, at least one user terminal can also perform a specific role in the various platform environments (e.g., a role such as a 'creator' participating in a project to generate data or a 'consumer' requesting data). At this time, the concept of 'role' in the present disclosure is a concept arbitrarily assigned according to the purpose of a specific entity, and is not intended to limit the invention to the name or number of the role.

[0073] Additionally, the platform providing device (110) may provide a data platform environment (150) by establishing a virtual environment (or virtual space) that can be used by at least one user terminal. In this case, the virtual environment in the present disclosure is a concept that encompasses virtual reality or augmented reality accessible through an AR device or VR device, or a virtual space on the Internet accessible through a mobile device, PC, etc.

[0074] For example, when at least one user terminal accesses a virtual space built by the platform providing device (110), the platform providing device (110) can transmit a virtual space built with various services and functions provided by the platform providing device (110) to the user terminal, and accordingly, users can use the various services and functions. In addition, for example, when at least one user terminal accesses a virtual space built by the platform providing device (110), the platform providing device (110) can be implemented to create at least one avatar corresponding to at least one user terminal so that the user can use the avatar to use various services and functions in the virtual space.

[0076] [Hardware Configuration]

[0077] FIG. 2 is a diagram illustrating the hardware configuration of an electronic device (e.g., a platform providing device or a server and / or administrator terminal included in the platform providing device) according to various embodiments.

[0078] Referring to FIG. 2, an electronic device (110) (or computing device, e.g., a server) may include various configurations for providing a data platform environment. Specifically, the electronic device (110) may include a memory (210) for storing data and various instructions to be transmitted to a processor, at least one processor (220) for performing operations based on instructions received from the memory (110) and controlling the electronic device (110), and a communication circuit (230) for enabling communication between the electronic device (110) and an external device. Furthermore, it is not limited thereto, but may include configurations such as hardware configurations and communication protocols typically included in data servers, etc.

[0079] Additionally, optionally, the electronic device (110) may further include an input device (not shown). In this case, the input device is a device that first receives user input from the outside, and for example, the electronic device (110) may further include at least one input device such as a keyboard or a mouse.

[0080] Additionally, optionally, the electronic device (110) may further include an output device (not shown). In this case, the output device is a device for displaying specific information externally from the at least one processor (220), and for example, the electronic device (110) may further include at least one output device such as a display, a VR device, AR glasses, an AR projector, or a printing device.

[0081] Additionally, the aforementioned internal components included in the electronic device (110) (e.g., memory, at least one processor, etc.) can transmit data to or communicate with each other through an internal bus (bus, 240).

[0083] Additionally, at least one processor (220) may include various configurations distinguished according to the functions provided. Specifically, at least one processor (220) may include, but is not limited to, a data generation unit (221) configured to provide a data generation environment to provide a function to generate synthetic data to a user, a data trading unit (222) configured to provide a data trading environment to provide a function to trade data to a user, and a data utilization unit (223) configured to provide a data utilization environment to provide a function to utilize data in various ways to a user. In particular, the data utilization unit (223) may be further divided into detailed configurations according to various uses in which the data can be utilized. Detailed information regarding the various data environments provided by at least one processor (220) is explained through the description of FIG. 3.

[0085] Additionally, the memory (210) may include various configurations that are arbitrarily (or hardware-wise) distinguished according to purpose and use. Specifically, the memory (210) may include an instruction storage unit (211) configured to store instructions that direct an operation to at least one processor (220), a synthetic data storage unit (212) configured to store synthetic data obtained through a data generation unit (221) and information associated therewith (e.g., information such as the quality of the synthetic data, domain, or reference data), an asset storage unit (213) configured to store block data containing data transaction information obtained through a data transaction unit (222) and information associated therewith (e.g., contribution, reward, or information about the connected blockchain network), and a model storage unit (214) configured to store information about a reference model or information associated therewith (e.g., reference model, etc.) generated through a data utilization unit (223), but the configuration of the memory is not limited to that described in FIG. 2, and the memory (210) may store various types of data generated or received by the electronic device (110).

[0087] [Various environments implemented on the platform]

[0088] FIG. 3 is a diagram illustrating various environments provided in a data platform environment provided by an electronic device according to various embodiments.

[0089] The term "environment" as defined in this disclosure refers to concepts such as a virtual reality environment implemented to enable a user to perform specific functions, or an internet environment accessible via an internet network.

[0090] Referring to FIG. 3, an electronic device (e.g., a platform providing device) according to one embodiment of the present disclosure may establish and provide a data platform environment (300). For example, the electronic device may establish and provide a virtual environment (e.g., a metaverse platform) in which a user can enjoy various user experiences using a virtual avatar. Additionally, for example, the electronic device may establish and provide a virtual environment accessible to the user through an immersive output device (e.g., augmented reality (AR) glasses, virtual reality (VR) devices, or extended reality (XR) devices). Additionally, for example, the electronic device may establish and provide an internet platform environment accessible to the user through a PC or mobile application. Thus, the data platform environment (300) implemented by the electronic device is not limited to a specific environment, but it is preferable to establish it as a virtual environment that induces multiple users to proactively perform various functions.

[0091] According to various embodiments, the electronic device may provide a data generation environment (310). In this case, the electronic device may provide various functions through the data generation environment (310).

[0092] For example, the electronic device may provide a project control function (311). In this disclosure, a project (or a virtual environment for data generation) may be defined as a data generation process or a unit in which a data generation process is performed, in which various entities participate in a platform environment to generate a data set that satisfies specific requirements. For example, various data providers (creators) may participate in the project and generate synthetic data using provided data generation tools, and the project may continue until the data set obtained based on the generated synthetic data satisfies specific requirements. The electronic device may provide a project control function (311), such as creating a project, determining the progress of a project, or determining whether to terminate a project.

[0093] A project-based data generation method performed by an electronic device is described in detail below (Figs. 8 to 17).

[0094] Additionally, for example, the electronic device may provide a creation tool provision function (312) for data creation. In this disclosure, the creation tool (or data creation tool) may refer to a tool (e.g., SW program, etc.) used by a data provider (creator) to create synthetic data in a platform environment. The electronic device may provide a creation tool provision function (312) that provides various creation tools stored in advance to a user of the platform environment.

[0095] The types of data generation tools provided by the electronic device are described in detail below (description of Fig. 7).

[0096] Additionally, for example, the electronic device may provide a visualization function (323) for various information performed in the data generation environment (310). Specifically, the electronic device may visualize and output the data generation environment (310) and the information associated with the data generation environment (310) to a user terminal.

[0097] Various UIs (User Interfaces) implemented as the data generation environment (310) is visualized on the user terminal are described in detail below.

[0099] According to various embodiments, the electronic device may provide a data trading environment (320). In this case, the electronic device may provide various functions through the data trading environment (320).

[0100] For example, the electronic device may provide a data transaction brokerage function (321). In this disclosure, data transaction is a concept that encompasses the exchange of data (e.g., synthetic data set generated in a data generation environment) between subjects in a platform environment, a trade involving rewards, and compensation for data generation. The electronic device may provide a data transaction brokerage function (321) that enables data trading by brokering data transactions between subjects.

[0101] Detailed information regarding the data transaction brokerage function (321) provided by the electronic device is described in detail below (Figs. 19 to 28).

[0102] Additionally, for example, the electronic device may provide a contribution calculation function (322). In the present disclosure, contribution may refer to the degree of contribution made by each subject of the platform environment to the generation of data (for example, a synthetic data set generated in a data generation environment). The electronic device may provide a contribution calculation function (322) that evaluates and quantifies the degree of contribution made by the subjects involved in the generation of the data set to the generation of the data set.

[0103] Detailed information regarding the contribution calculation function (322) provided by the electronic device is described in detail below.

[0104] Additionally, for example, the electronic device may provide a visualization function (323) for various information performed in a data trading environment. Specifically, the electronic device may visualize and output the data trading environment (320) and the information associated with the data trading environment (320) to a user terminal.

[0105] Various UIs (User Interfaces) implemented as the data transaction environment (320) is visualized on the user terminal are described in detail below.

[0107] According to various embodiments, the electronic device may provide a data utilization environment (330). In this case, the electronic device may provide various functions through the data utilization environment (330).

[0108] For example, the electronic device may provide an AI model providing function (331). Specifically, the electronic device may provide an AI model providing function (331) such as training a reference model stored in advance based on a data set (e.g., a synthetic data set generated in a data generation environment), distributing a trained AI model, verifying an AI model based on test data received from an external source, or providing feedback to a data provider based on the verification results.

[0109] Additionally, for example, the electronic device may provide a preview function (332). In this disclosure, a preview may mean exhibiting or showing an example of applying a data set (e.g., a synthetic data set generated in a data generation environment) or an AI model obtained using the data set to an actual task in a specific domain. The electronic device may provide a preview function (332), such as visualizing and exhibiting the generated data set or demonstrating the generated AI model.

[0110] Additionally, for example, the electronic device may provide a visualization function (333) for various information performed in a data utilization environment. Specifically, the electronic device may visualize and output the data utilization environment (330) and the information associated with the data utilization environment (330) to a user terminal.

[0111] Various UIs (User Interfaces) implemented as the data utilization environment (330) is visualized on the user terminal are described in detail below.

[0113] [Terminal Configuration]

[0114] FIG. 4 is a diagram illustrating the hardware configurations of an electronic device (e.g., a user terminal such as a mobile phone, PC, etc.) according to various embodiments.

[0115] An electronic device (120) according to one embodiment of the present disclosure may mean a user terminal for using a data platform environment provided by a server, but is not limited thereto.

[0116] Referring to FIG. 4, an electronic device (120) according to one embodiment of the present disclosure may include a memory (410), at least one processor (420), a communication circuit (430), and an output device (440). In this case, a general description of the memory (410), at least one processor (420), and the communication circuit (430) is omitted.

[0117] The output device (440) may include, for example, a general display as well as the aforementioned immersive output device (e.g., AR (augmented reality) glasses, VR (virtual reality) devices, and XR (extended reality) devices). Additionally, the output device (440) may have the function of an input device implemented together to enable communication with the user. At least one processor (420) may output a virtualized platform environment through the output device (440). For example, when a specific input is received by a user in the virtualized space implemented by the output device (440), the electronic device may provide a corresponding user interaction.

[0118] The user interface environment output by the output device (440) is described below.

[0120] FIG. 5 is a diagram illustrating examples in which a data platform environment is implemented through an electronic device according to various embodiments.

[0121] An electronic device according to one embodiment of the present disclosure can output a data platform environment using an output device.

[0122] For example, referring to FIG. 5(a), the electronic device (120) can output a virtual space (500) in which various environments of a data platform according to an embodiment are implemented. At this time, the virtual space (500) may be implemented as a two-dimensional or three-dimensional environment, but is not limited thereto.

[0123] Additionally, the electronic device (120) can output an avatar (501) corresponding to a user in the virtual space (500). At this time, the avatar (501) may be a computer graphic icon representing the user in the virtual space (500). At this time, the characteristics of the avatar (501) may be set by the user or arbitrarily set. For example, characteristics such as the appearance or attire of the avatar (501) may be selected by the user or automatically set by the system. Additionally, the avatar (501) may move to various spaces in the virtual space (500) according to the user's command.

[0124] Additionally, the electronic device (120) can output a plurality of sub-spaces (510, 520, 530) corresponding to a plurality of environments implemented on a data platform in the virtual space (500). For example, the electronic device (120) can output a first sub-space (510) corresponding to a data generation environment, a second sub-space (520) corresponding to a data trading environment, and a third sub-space (530) corresponding to a data utilization environment. That is, a user can access an environment corresponding to a sub-space by entering one of the plurality of sub-spaces (e.g., access via an avatar or input to a sub-space).

[0125] As another example, referring to FIG. 5(b), the electronic device (120) can output a screen to a first area (503) of the output device for accessing at least one of the various environments of the data platform according to the embodiment through the output device, and can output a screen corresponding to the environment selected through the first area to a second area (505) of the output device.

[0126] Additionally, the electronic device (120) may display a plurality of icons (504a, 504b, 504c) corresponding to a plurality of environments of the data platform in the first area (503). For example, the electronic device (120) may output a first icon (504a) corresponding to a data generation environment, a second icon (504b) corresponding to a data trading environment, and a third icon (504c) corresponding to a data utilization space.

[0127] Additionally, the electronic device (120) may provide an environment corresponding to an icon selected by the user among the plurality of icons through the second area (505). For example, if the first icon (504a) is selected by user input, the electronic device (120) may output a data generation environment corresponding to the first icon (504a) to the second area (505). More specifically, the electronic device (120) may output a list of currently ongoing data generation projects (e.g., AA data set generation project, BB data set generation project, CC data set generation project, etc.), but is not limited thereto, and may output a list of projects scheduled to proceed, a list of completed projects, a list of projects available for participation, a list of projects not available for participation, etc.

[0129] [Various operations performed by electronic devices in a data platform environment]

[0130] An electronic device may be assigned at least one role depending on the operation or purpose performed within the data platform environment. Specifically, the electronic device may assign at least one role to entities using the data platform environment based on pre-stored role information assigned according to the type of data platform environment. For example, in a data generation environment, an electronic device controlling a project (e.g., a server, an administrator terminal, etc.) may be assigned the role of a 'producer' that designs the project and requests the generation of data. Additionally, for example, in a data generation environment, an electronic device generating data (e.g., a user terminal) may be assigned the role of a 'creator' that participates in the project to generate data. Furthermore, for example, in a data generation environment, an electronic device requesting data (e.g., a user terminal) may be assigned the role of a 'consumer' that requests data satisfying specific requirements. Additionally, for example, in a data generation environment, an electronic device inspecting generated data may be assigned the role of a 'checker' that verifies the generated data and determines whether to terminate the project. Furthermore, for example, in a data trading environment, an electronic device purchasing data may be assigned the role of a 'buyer'. Also, for example, in a data trading environment, an electronic device selling data may be assigned the role of a 'seller'.

[0131] In addition, multiple roles may be assigned to a single electronic device. Specifically, the electronic device may be assigned multiple roles in the same environment (e.g., cases where multiple roles are assigned within a single project in a data generation environment, or cases where different roles are assigned for each project), or may be assigned multiple roles in different environments. For example, a server providing a data platform environment may be assigned the role of a 'producer' who designs a project and the role of an 'inspector' who verifies generated data and determines whether to terminate the project in a data generation environment, and may be assigned the role of a 'seller' who sells data in a data trading environment, but is not limited thereto. That is, the embodiments of the present disclosure are not limited by the names of the roles assigned to the electronic device, and a role with a name corresponding to the operation may be assigned according to the operation of the subject within the platform environment.

[0132] FIG. 6 is a diagram illustrating operations performed by an electronic device in a data platform environment according to various embodiments.

[0134] * Referring to FIG. 6(a), at least one electronic device connected to a platform server can perform a data generation operation. Specifically, a first electronic device (610, e.g., a user terminal acting as a 'consumer') can request a second electronic device (620, e.g., a server acting as a 'producer') to generate a data set (S601). In this case, the second electronic device (620) can generate a project (600) for generating the data set based on the reception of the request (S603).

[0135] Additionally, the second electronic device may determine at least one electronic device (e.g., a third electronic device (630a) and / or a fourth electronic device (630b)) as the entity performing the project (600). The third electronic device (630a) and / or the fourth electronic device (630b) determined as the entity performing the project (600) may generate data using a data generation tool (640). At this time, the data generation tool (640) is a tool for generating synthetic data and may be stored in advance on a platform server. (Details regarding the data generation tool are described in FIG. 7.) Specifically, the server may generate synthetic data using the data generation tool (640) for the third electronic device (630a) and the data generation tool (640) for the fourth electronic device (630b), and the platform server may generate a synthetic data set (650) based on at least a portion of the generated synthetic data.

[0136] Detailed information regarding data generation operations is explained below.

[0137] Referring to FIG. 6(b), at least one electronic device connected to a platform server can perform data trading operations. Specifically, a fourth electronic device (660, e.g., a user terminal and / or platform provider device acting as a 'buyer') and a fifth electronic device (670, e.g., a user terminal and / or platform provider device acting as a 'seller') can perform data trading operations in a data trading environment. For example, the fourth electronic device (660) can receive a synthetic data set (650) generated from the fifth electronic device (670) by paying a predetermined reward (680) to the fifth electronic device (670).

[0138] Detailed information regarding data trading operations is explained below.

[0139] Referring to FIG. 6(c), at least one electronic device connected to the platform server can perform data utilization operations. For example, at least one electronic device can train or evaluate an AI model (690) using the generated synthetic data set (650), but is not limited thereto.

[0140] Detailed information regarding data utilization operations is explained below.

[0142] [Data Generation Environment]

[0143] FIG. 7 is a diagram illustrating information that can be generated or received by an electronic device and stored in memory according to various embodiments.

[0144] An electronic device according to various embodiments of the present disclosure may generate a project for generating a data set (e.g., a synthetic data set). Specifically, the electronic device may generate at least one project information including various information related to the generation of the data set and store it in memory (210). For example, the electronic device may store a first project information (710) currently in progress, a second project information (720), and a third project information (730) that has been completed in memory (210).

[0145] Additionally, the electronic device may store information in different areas of the memory (210) based on the current progress status of the project. For example, the electronic device may store information on a project currently in progress in a first memory area (700a, e.g., a memory area for continuous updating), and information on a completed project or a project scheduled for progress in a second memory area (700b, e.g., a memory area where updating is unnecessary or data is deleted after a certain period of time).

[0146] Additionally, the project information generated by the electronic device may include at least one metadata related to the project. Specifically, the at least one metadata corresponding to the project information may include information related to the data set to be generated through the project, information related to the progress of the project, or attribute information of the project. For example, the first project information (710) may include, but is not limited to, level information (711) indicating the difficulty of generating the data set to be generated through the first project, participant information (712) regarding the entity participating in the first project, generation tool information (713) related to the generation tool used to generate the data set through the first project, request information (714) related to the data generation conditions of the first project, reference information (715) indicating reference data similar to the data set to be generated through the first project, and generation information (716) indicating the current generation status of the data set through the first project.

[0147] Additionally, at least one metadata corresponding to the project information may be updated. For example, the generation information (716) included in the first project information (710) may be updated according to the degree of generation of the data set. For another example, the participant information (712) included in the first project information (710) may be updated according to the number of subjects determined to participate in the first project. For yet another example, the request information (714) included in the first project information (710) may be updated according to a request received from a data user.

[0148] Additionally, for example, the first metadata included in the project information may be updated based on the second metadata included in the project information. As a specific example, the level information (711) included in the first project information (710) may be updated according to the degree of generation of the data set confirmed by the generation information (716). Specifically, the electronic device may be configured such that the difficulty of the project decreases as the completeness of the data set generated through the project increases, but is not limited thereto.

[0149] In addition, for example, the generation tool information (713) included in the first project information (710) can be determined based on the data generation tool information (750).

[0150] An electronic device according to one embodiment of the present disclosure may store in advance in memory (210) data generation tool information (750) used to generate a data set (e.g., a synthetic data set). For example, the data generation tool information (750) may include, but is not limited to, a generative model (751), a neural rendering model (752), a computer graphics (CG) tool (753) for generating synthetic data by modeling and rendering images, etc. using a computer, a shooting tool (754) for generating synthetic data based on images captured using a specific device (e.g., a general camera, an AR or VR camera, etc.), a domain-specific tool (755) for generating synthetic data specialized for a specific domain, and a data collection tool (756) for obtaining data based on user input (e.g., data on the user's preference (priority) for a specific image) using a gamification tool, etc.

[0151] At this time, the generative model (751) is a model for learning input data and generating similar data that follows the distribution of the input data. For example, an electronic device may provide a supervised generative model (such as Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), etc.) or an unsupervised generative model (for example, a statistical generative model such as Kernel Density Estimation, or a deep learning-based generative model such as Generative Adversarial Network (GAN) or Variational Auto-Encoder (VAE)), and a user terminal may generate a synthetic data set using the generative model (751).

[0152] Additionally, the neural rendering tool (752) can operate by training a neural renderer that learns hidden information of a three-dimensional scene through multiple two-dimensional rendering images, and then using the trained neural renderer to infer a two-dimensional rendering image corresponding to a condition that did not appear in the training.

[0153] Additionally, the CG tool (753) can operate by acquiring virtual computer graphics data by modeling and rendering graphics based on at least one parameter (e.g., amount of light, angle of light, etc.).

[0154] Additionally, the shooting tool (754) may operate in a manner that generates composite data based on image data obtained by an electronic device including a shooting device such as a camera. For example, a method for generating a composite data set using the shooting tool (754) may include the operation of the electronic device obtaining a composite data set based on at least one image captured by a user terminal and an image processed from the at least one image.

[0155] Additionally, the domain-specific tool (755) may be a generation tool designed to be optimized for the domain of the data. For example, an electronic device may provide at least one data provider with a tool for generating waste plastic images (e.g., waste plastic CG data, etc.) to generate a synthetic data set of waste plastic domains.

[0156] Additionally, the data collection tool (756) may operate in a manner that acquires data related to the user's preferences based on user communication. For example, the electronic device may display at least two images on a user terminal and acquire data related to the user's preferences based on an image selected by the user among the at least two images.

[0157] An electronic device according to one embodiment of the present disclosure may generate and store a matching table (770) based on the association between various pieces of information stored in memory (210). Specifically, the electronic device may generate a matching table (770) based on the association between a data generation tool and an attribute of a data set (e.g., data domain, modality, etc.). For example, the electronic device may generate a matching table corresponding to an optimal data generation tool (CG tool) for generating data of a corresponding domain according to the domain of the data set (e.g., data regarding images of waste plastic), and store it in memory (210). In addition, in this case, when a project for generating a specific data set is created, the electronic device may identify a data generation tool corresponding to the domain of the data set to be generated based on the matching table (770). The electronic device may provide the identified data generation tool to at least one electronic device participating in the project (e.g., a user terminal acting as a supplier).

[0159] FIG. 8 is a flowchart illustrating a method in which at least one electronic device generates a data set according to various embodiments.

[0160] FIG. 9 is a diagram illustrating, according to various embodiments, an exemplary method in which at least one electronic device generates a data set.

[0161] An electronic device according to one embodiment of the present disclosure may generate synthetic data at each of at least one electronic device and generate a data set based on the generated synthetic data. Specifically, the electronic device may generate a data set by creating a project for generating a data set, involving at least one electronic device in the project, and terminating the project when the generation of the data set by at least one electronic device is completed.

[0162] For example, referring to FIG. 8, a first electronic device may receive a request for the creation of a data set (S810). At this time, the first electronic device may receive the request from a second electronic device (e.g., a user terminal acting as a 'consumer'). For example, referring to FIG. 9, a consumer terminal (901) may transmit a request for the creation of a data set to a server (902). Specifically, the request for the creation of the data set may include requirements for the data set to be created. For example, the request for the creation of the data set may include, but is not limited to, properties of the data set such as the domain, volume, or modality of the required data set, the quality of the required data set, or sample data for the required data set.

[0163] Additionally, the first electronic device can identify a reference data set based on the reception of a request for the creation of a data set (S830). At this time, the first electronic device can identify the attributes of the required data set based on the request for the creation of the data set, and can search for at least one reference data set corresponding to the attributes of the identified data set among a plurality of reference data sets stored in advance. For example, referring to FIG. 9, the server (902) can identify a first reference data set (910) corresponding to a request received from a user terminal (901) among reference data sets stored in advance in a pre-built database (905, or memory).

[0164] Additionally, the first electronic device may create a first project by creating first request information that includes a first creation condition set based on a relationship with the reference data set (S850). Specifically, the first electronic device may acquire first request information that includes a first creation condition set based on a relationship with the reference data set, and may provide a first virtual environment for data creation based on the first request information. At this time, providing the first virtual environment (or first project) may mean allowing access to the first virtual environment on the platform to multiple platform users. At this time, the first electronic device may create and store in memory first project information that includes all information associated with the first project, including the first request information. Specifically, the first request information may include the first creation condition, the confirmed reference data set, or sample data included in the received request. For example, referring to FIG. 9, the server (902) can create a first project (900) (or a first virtual environment) by creating first request information including first creation conditions set based on a relationship with a first reference data set (910). Additionally, the server (920) can transmit the first creation conditions to a first provider terminal (903a) participating in the first project (900).

[0165] At this time, the first electronic device may determine, according to predetermined criteria, at least one electronic device that will participate in the first project for data generation and generate data. Specifically, when the first electronic device receives an intention to participate in the project (or a request for access to the first virtual environment) from at least one electronic device, it may determine whether said at least one electronic device is suitable for performing the project. For example, the first electronic device may determine whether said at least one electronic device will participate in the project by determining, based on the level information of said first project, whether the level of data generation by said at least one electronic device corresponds to said level information. Furthermore, not limited thereto, the first electronic device may select at least one electronic device suitable for participating in the project and generating data based on the first project information and transmit a suggestion for project participation to said at least one electronic device. If the level information of the project is very low (e.g., data generation through the selection of preferred images), all electronic devices may be configured to participate in the project regardless of their data generation level.

[0166] Additionally, the first request information generated by the first electronic device may include at least one additional generation condition in addition to the first generation condition. Specifically, the electronic device may generate multiple generation conditions that are different from each other based on a request received from a user terminal (901) and a confirmed first reference data set (910), and may transmit the multiple generation conditions to multiple supplier terminals. For example, the server (902) may transmit the first generation condition to the first supplier terminal (903a) and transmit a second generation condition that is different from the first generation condition to the second supplier terminal (903b). At this time, the relationship between the first generation condition and the second generation condition transmitted to at least one supplier terminal and the first reference data set (910) may be set differently. Specifically, the server (902) may generate a first generation condition requesting the generation of data having a first association relationship with the first reference data set (910) and transmit it to the first provider terminal (903a), and may generate a second generation condition requesting the generation of data having a second association relationship different from the first association relationship with the first reference data set (910) and transmit it to the second provider terminal (903b). At this time, the association relationship with the first reference data set (910) may be based on an Image of Data. A method for providing an Image of Data-based generation condition is described in detail in FIG. 11.

[0167] Additionally, the first electronic device may transmit the same generation conditions to different electronic devices, but is not limited thereto. For example, the server (902) may transmit the second generation conditions to the second provider terminal (903b) and may transmit the third generation conditions, which are identical to the second generation conditions, to the third provider terminal (903c).

[0169] Additionally, the first electronic device may provide at least one data generation tool (to the data provider). Specifically, the first electronic device may select at least one of a plurality of data generation tools stored in memory based on predetermined criteria and provide it to the data provider. For example, the first electronic device may select and provide at least one generation tool suitable for generating the required data based on a matching table. Additionally, for example, the first electronic device may provide at least one data generation tool used to generate the reference data set based on the identified reference data set. Additionally, for example, the first electronic device may provide at least one data generation tool identified by the request based on the request received from the user. Additionally, for example, the first electronic device may select and provide at least one data generation tool suitable for the data generation conditions. Accordingly, at least one electronic device participating in a project for data generation (e.g., a terminal acting as a data provider) may generate data using the provided at least one data generation tool.

[0170] Referring again to FIG. 9, the server (902) may provide a first data generation tool (920a) to a first provider terminal (903a), and the first provider terminal (903a) may generate a first data set (930a) using the first data generation tool (920a).

[0171] Additionally, the server (902) may provide different data generation tools to multiple provider terminals. For example, the server (902) may provide a first data generation tool (920a) to a first provider terminal (903a), a second data generation tool (920b) to a second provider terminal (903b), and a third data generation tool (920c) to a third provider terminal (903c). At this time, the data generation tool provided to each provider terminal may be selected considering the level of data generation, specialized field, required contribution, etc. of the provider terminal, but is not limited thereto. Details regarding this are explained in FIG. 11.

[0173] The first electronic device may provide a data imaging tool for providing an image of data corresponding to the generated data. In this case, the data image may be an image that visualizes the data by reflecting the distribution of the data. Specifically, the data image may be a visual image that reflects the intrinsic relationships of the data (e.g., distance between data, distribution of data, bias of data, etc.), and the data may appear as individual data points on the data image. That is, by providing a data image of the data generated through the data imaging tool, the first electronic device may induce the data provider to check the attributes of the data currently being generated (e.g., distribution of data, density, relationship with other data, etc.).

[0174] Referring again to FIG. 9, the server (902) may provide a data imaging tool (940) and may provide a reference image (950) corresponding to a reference data set (910) verified based on the data imaging tool (940). Additionally, the first provider terminal (901a) may verify a first data image (960a) corresponding to a generated first data set (930a) using the data imaging tool (940). At this time, the first data image (960a) may be provided so that the data generated by the first provider terminal (901a) is visually highlighted, but is not limited thereto. Additionally, the second provider terminal (902b) and / or the third provider terminal (902c) can use the data imaging tool (940) to verify the second data image (960b) corresponding to the generated second data set (930b) and / or the third data set (930c). Additionally, the server can provide the third data image (960c) corresponding to the synthetic data set (970) which includes the reference data set and the data sets generated by all data providers.

[0175] In addition, the first electronic device can provide a visual indication corresponding to the data generation conditions along with the data image appearing using a data generation tool. Details regarding this are explained in FIG. 11.

[0177] Additionally, referring again to FIG. 8, the first electronic device can determine whether a synthetic data set containing data generated from at least one electronic device participating in the first project satisfies the first generation condition (S870). Specifically, the first electronic device can check whether a synthetic data set generated from at least one user terminal accessing the first virtual environment satisfies the first generation condition. Additionally, if the synthetic data set satisfies the first generation condition, the first electronic device can terminate the first project (S890). Specifically, based on the determination result that the synthetic data set satisfies the first generation condition, the first electronic device can block user access to the first virtual environment. That is, to determine whether to terminate the project for data generation, the first electronic device can determine whether the data generated by at least one electronic device meets the generation condition associated with the project.

[0178] In addition, if multiple generation conditions are assigned to multiple electronic devices participating in the project, the first electronic device may determine whether to terminate the project based on the multiple generation conditions. For example, the first electronic device may decide to terminate the project if conditions among the multiple generation conditions that meet or exceed a predetermined standard are satisfied, but is not limited thereto.

[0179] Referring again to FIG. 9, the server (902) can determine whether to terminate the first project (900) for data generation by determining whether a synthetic data set (970) containing data generated by a plurality of supplier terminals satisfies the data generation conditions. Specifically, the server (902) can determine whether the synthetic data set (970) satisfies a first generation condition requiring the generation of data having a first association relationship with the reference data set (910), a second generation condition requiring the generation of data having a second association relationship with the reference data set (910), and a third generation condition requiring the generation of data having a third association relationship with the reference data set (910), above a predetermined standard. For example, the server (902) can terminate the project for data generation if the synthetic data set (970) satisfies all of the plurality of generation conditions. Additionally, for example, the server (902) may terminate the project for data generation if the synthetic data set (970) satisfies a certain ratio or more of the plurality of generation conditions. As a specific example, the server (902) may determine whether the plurality of generation conditions satisfy a predetermined standard or more by checking the ratio of data generated in a specific area where data generation is required, based on a third data image (960c) corresponding to the synthetic data set (970).

[0181] FIG. 10 is a diagram illustrating the operation of an electronic device updating information of a project for data generation according to various embodiments.

[0182] Referring to FIG. 10, when an electronic device receives a request for data generation, it can generate request information (1010) based on the request and store it in memory (210). At this time, the request information (1010) may include at least some of a generation condition (1011) associated with the data generation requirement, sample data (1013) received from a data user, a reference data set (1015), or a target image (1017) generated by virtually generating an imaging result of the data set to be generated.

[0183] Additionally, the electronic device may store reference information (1030) (e.g., a database (DB) built based on a synthetic data set generated within a platform) including various reference data sets associated with the synthetic data set in memory (210).

[0184] Additionally, the electronic device can generate generation information (1050) containing information regarding the generated synthetic data and store it in memory (210).

[0186] According to various embodiments, the electronic device may perform a reflection operation (1060) to generate request information (1010) based on reference information (1030). Specifically, the electronic device may generate the request information (1010) based on the reference information (1030) according to the reflection operation (1060). For example, the electronic device may obtain the request information (1010) by including a first reference data set (1031) included in the reference information (1030). In this case, the request information (1010) may include, but is not limited to, the first reference data set (1015), as well as a generation condition defined based on similarity with the first reference data set (1031) or a target image in which the first reference data set (1031) is reflected. The reflection operation (1060) may be performed in the step of creating a project for generating data.

[0187] Additionally, the electronic device may perform a comparison operation (1070) that compares request information (1010) and generation information (1050) to determine whether the required data and the generated data correspond. Specifically, the electronic device may compare a synthetic data set (1051) included in the generation information (1050) with a generation condition (1011) included in the request information (1010) to determine whether the generated data meets the required conditions. For example, the electronic device may perform the comparison operation (1070) by evaluating the quality of the generated synthetic data set (1051) to determine whether the quality required by the generation condition (1011) has been achieved, but is not limited thereto, and may perform the comparison operation (1070) by checking a plurality of elements included in the generation condition (1011) (e.g., attributes, generation parameters, similarity with sample data, similarity with reference data). The above comparison operation (1070) can be performed during the process of a project for data generation.

[0188] To evaluate the quality of the synthetic data set described above, the electronic device may store and use a quality evaluation model in advance. Specifically, the electronic device may calculate the distance between data included in the synthetic data set and evaluate the quality of the synthetic data set based on the said distance. For example, the electronic device may calculate an evaluation index representing the quality of the synthetic data set based on at least one of the density, bias, or uniformity between data of the synthetic data set identified based on the distance values ​​between the data.

[0189] Additionally, the electronic device may perform an update operation (1080) to update reference information (1030) based on the generated information (1050). Specifically, the electronic device may update the reference information (1030) by storing the generated synthetic data set as a reference data set in a database. For example, the electronic device may obtain a second reference data set (1033) based on a synthetic data set generated based on a first synthetic data set, a second synthetic data set, and a third synthetic data set, and update the reference information (1030) to include the second reference data set (1033). At this time, the update operation (1080) may be performed after the project for data generation has ended.

[0191] FIG. 11 is a diagram illustrating a method of performing a data generation operation by providing a data image to an electronic device according to various embodiments.

[0192] Referring to FIG. 11, the electronic device can acquire a reference image (1120) based on a reference data set (1110) using a data imaging tool (1150). At this time, the reference image (1120) may be data that visualizes the data included in the reference data set (1110) as data points.

[0193] Additionally, the electronic device may generate a reference image (1120) including a visual display corresponding to at least one generation condition associated with the required data. Specifically, the electronic device may generate a reference image (1120) including a first visual display (1121) corresponding to a first generation condition (e.g., data having a first association relationship with the reference data set), a second generation condition (e.g., data having a second association relationship with the reference data set), and a third generation condition (e.g., data having a third association relationship with the reference data set). In this case, the electronic device may provide the visual display to an area where data is required on an imaging space where the data image is defined. For example, the electronic device may generate the first visual display (1121) to represent a first area of ​​the imaging space, generate the second visual display (1123) to represent a second area of ​​the imaging space, and generate the third visual display (1125) to represent a third area of ​​the imaging space. As a result of providing a visual guide for data generation in this manner, the data provider terminal may be required to generate data corresponding to the area corresponding to the visual display.

[0194] Additionally, the electronic device may provide a data generation tool corresponding to at least one generation condition to at least one supplier terminal. Specifically, the electronic device may select and provide the most suitable data generation tool for generating data that aligns with the generation condition based on the at least one generation condition. For example, the electronic device may provide a first generation tool (1130a) to a first data supplier to request data generation according to a first generation condition, provide a second generation tool (1130b) to a second data supplier to request data generation according to a second generation condition, and provide a third generation tool (1130c) to a third data supplier to request data generation according to a third generation condition. Accordingly, the first data provider can generate a first synthetic data set (1140a) using the first generation tool (1130a), the second data provider can generate a second synthetic data set (1140b) using the second generation tool (1130b), and the third data provider can generate a third synthetic data set (1140c) using the third generation tool (1130c).

[0195] In addition, the electronic device may provide additional information related to data generation together with the data generation tool. For example, while providing a CG tool, the electronic device may also provide a range of CG parameters that meet the generation conditions, but is not limited thereto.

[0196] Additionally, the electronic device can acquire a data image corresponding to a generated synthetic data set by using a data imaging tool (1150). For example, the electronic device can acquire a first data image (1160a) based on a first synthetic data set (1140a), acquire a second data image (1160b) based on a second synthetic data set (1140b), and acquire a third data image (1160c) based on a third synthetic data set (1140c) by using the data imaging tool (1150). In this case, the electronic device can display the data included in the synthetic data set generated by the data provider as visually different from the existing data.

[0197] Additionally, the electronic device can acquire a combined data image (1170) corresponding to a synthetic data set based on the plurality of synthetic data sets (1140a, 1140b, 1140c) or a plurality of data images (1160a, 1160b, 1160c).

[0198] In this case, the electronic device can determine whether a plurality of data points included in the aggregated data image (1170) have been generated in an area required by at least one generation condition. For example, the electronic device can determine whether a plurality of data points corresponding to a first synthetic data set (1140a) included in the aggregated data image (1170) have been generated in a first area of ​​the imaging space corresponding to the first generation condition. Accordingly, the electronic device can determine whether the synthetic data set generated from data providers has been generated in accordance with the generation condition.

[0200] FIG. 12 is a drawing illustrating a user interface in which an electronic device generates data using a data generation tool according to various embodiments.

[0201] Referring to FIG. 12, an electronic device (e.g., a user terminal) can display various information by configuring a user interface for generating data using an output device (1200). At this time, the output device (1200) may be a configuration included in the electronic device or connected via electricity and / or communication, and as described above, may include not only a general display screen but also a virtual display screen such as virtual reality and / or augmented reality.

[0202] Specifically, the output device (1200) may include a first space (1210) for selecting data generation tools used to generate data. The electronic device may display tool information (1215) containing data generation tools in the first space (1210) of the output device. At this time, at least some of the various data generation tools may be selected by the user, and the electronic device may be configured to provide at least one selected generation tool.

[0203] Additionally, the output device (1200) may include a second space (1220) for verifying general attributes regarding the data to be generated. The electronic device may display data attribute information (1225), such as the domain of the data and the modality of the data, in the second space (1220) of the output device. At this time, the data attribute information (1225) may be obtained based on request information for data generation. Additionally, the user may verify the attributes of the data to be generated through the data attribute information (1225).

[0204] Additionally, the output device (1200) may include at least one space for providing generation assistance information corresponding to each of the selected at least one generation tool. For example, the output device (1200) may include a third space (1230) for providing first generation assistance information (1235) corresponding to a first generation tool (e.g., CG). Additionally, for example, the output device (1200) may include a fourth space (1240) for providing second generation assistance information (1245) corresponding to a second generation tool (e.g., a domain-specific tool).

[0205] At this time, the generation assistance information may be obtained based on at least one generation parameter corresponding to the generation tool. Specifically, the electronic device may generate and display the generation assistance information based on at least one generation parameter corresponding to the generation tool selected by the user, and some of the at least one generation parameter may be adjusted based on user input regarding at least one generation parameter included in the third space (1230) and / or the fourth space (1240), and the user may check the generation parameters of the data currently being generated through the third space (1230) and / or the fourth space (1240).

[0206] Additionally, the generation aid information displayed through the output device may be determined based on the selected generation tool. Specifically, the electronic device may identify the selected generation tool, identify at least one generation parameter matching the identified generation tool, and generate and provide generation aid information to adjust or verify the at least one generation parameter. For example, the electronic device may identify at least one generation parameter specialized for "waste plastic" data based on user input for a domain-specific tool (e.g., a parameter specialized for the waste plastic domain that appears according to various light characteristics, not just RGB, such as contamination level, degree of wrinkling, transparency, shape, material, presence of a lid, color, size, and whether a label is attached), and generate generation aid information based on this and display it in the fourth space (1240).

[0207] Additionally, the output device (1200) may include a fifth space (1250) for displaying synthetic data generated (or being generated) using a selected generation tool. Specifically, the electronic device may display visual information (1255) of the synthetic data and / or a data image (not shown) corresponding to the synthetic data through the fifth space (1250) based on the synthetic data generated using the selected generation tool.

[0209] FIG. 13 is a flowchart illustrating a method of visualizing and providing data generated by an electronic device according to various embodiments.

[0210] FIG. 14 is a drawing illustrating a user interface that visualizes and provides data generated by an electronic device according to various embodiments.

[0211] Referring to FIG. 13, the electronic device can obtain a reference data set based on request information for data generation (S1310). Additionally, the electronic device can display a reference image visualized in an imaging space using an output device (S1320). For example, referring to FIG. 14, the electronic device can obtain a reference image (1410) based on a reference data set corresponding to the request information and display the reference image (1410) in a first space (1401) of the output device.

[0212] Additionally, the electronic device can generate a first synthetic data set according to user input using a data generation tool (S1330). At this time, the data generation tool may be selected by user input as described in the description of FIG. 12, but is not limited thereto; the electronic device may also select a generation tool according to a predetermined method based on the attributes of the data for which generation is requested and provide it to the user terminal. For example, referring to FIG. 14, the electronic device may display the first synthetic data set (1420) generated using the data generation tool in a second space (1402) of the output device. At this time, the second space (1402) may be configured as the user interface shown in FIG. 12, but is not limited thereto.

[0213] Additionally, the electronic device may display a first data image visualizing the first synthetic data set and the reference data set in the imaging space (S1340). Specifically, the electronic device may obtain the first data image by visually displaying the reference data set and the first synthetic data set differently (e.g., different colors, etc.). For example, referring to FIG. 14, the electronic device may obtain a first data image (1430) by displaying the reference data set (1410) and the first synthetic data set (1420) in a virtual imaging space, and may display the first data image (1430) in a third space (1403) of the output device.

[0214] Additionally, the electronic device may additionally or optionally display a second data image visualizing the first synthetic data set, the reference data set, and the additionally received second synthetic data set in the imaging space. In this case, the second synthetic data set may be data additionally generated from the terminal from which the first synthetic data set was received, or data generated from another terminal.

[0216] FIG. 15 is a flowchart illustrating information that an electronic device may additionally provide in relation to data generation according to various embodiments.

[0217] FIG. 16 is a drawing illustrating an embodiment of a user interface in which an electronic device provides additional information regarding data generation according to various embodiments.

[0218] FIG. 17 is a drawing illustrating another embodiment of a user interface in which an electronic device provides additional information regarding data generation according to various embodiments.

[0219] Referring to FIG. 15, after generating a first synthetic data set according to step S1330 of FIG. 13, the electronic device may generate and provide goal achievement information associated with whether the generation conditions included in the request information are satisfied (S1510). Specifically, the goal achievement information may include the degree to which the generation conditions confirmed by the request information are achieved or the degree to which the required data is matched. For example, the electronic device may provide a visual representation of the degree to which the synthetic data set generated by the data provider matches the required generation conditions.

[0220] For example, referring to FIG. 16, the electronic device may display a reference image (1610) corresponding to a reference data set and a first data image (1620) corresponding to the reference data set and the first synthetic data set using an output device (1600). At this time, the electronic device may generate and provide a visual target image (1650) for the synthetic data set to be generated. At this time, the target image (1650) may be a data image corresponding to the synthetic data set to be generated. The electronic device may generate and provide the target image (1650) by inferring data points that need to be generated according to request information. Additionally, the electronic device may generate target achievement information (1670) indicating the degree to which the first synthetic data set meets the data generation conditions and display it through the output device (1600). At this time, the electronic device may generate the target achievement information (1670) by visualizing the achievement rate in the form of a bar, but is not limited thereto. Additionally, if the generated first synthetic data set contains outlier data (1625) that does not meet the generation conditions, the electronic device may further provide a visual indication to induce the user to check the outlier data (1625). Additionally, not limited thereto, the electronic device may request the user terminal to improve the outlier data (1625) or provide improved data.

[0222] Additionally, optionally, additionally, or alternatively, after generating the first synthetic data set according to step S1330 of FIG. 13, the electronic device may generate and provide information on participants participating in data generation and information on contribution related to the degree of data generation by the participants (S1530). Specifically, the participant information may refer to information about data providers participating in the data generation project, and the contribution information may refer to information indicating the generation contribution of each provider determined based on the data generated by each data provider.

[0223] For example, referring to FIG. 17, an electronic device can display a data image (1710) of synthetic data being generated through a data generation project using an output device (1700). At this time, the electronic device can display a data image (1710) that includes a visual effect (1715) for displaying data generated by a user terminal connected to the output device (1700). Additionally, the electronic device can display participant information (1720) indicating information about data providers participating in the data generation project using the output device (1700). Additionally, the electronic device can display contribution information (1730) indicating the degree to which each data provider contributed to data generation using the output device (1700). In this case, the electronic device can provide a visual effect (1750) for displaying the participant corresponding to the user terminal connected to the output device (1700) and the contribution of that participant.

[0225] [Data Trading Environment]

[0226] An electronic device according to various embodiments of the present disclosure can provide a trading environment for data sets (e.g., synthetic data sets) among various entities. In the case of data trading through a platform, it is necessary to provide a secure trading environment by storing information related to the transaction in a decentralized manner, and also to establish criteria for equitably distributing rewards associated with the transaction among data providers who generated the data.

[0227] To provide a safe trading environment with clear reward distribution, an electronic device according to various embodiments of the present disclosure provides a data trading method based on a blockchain network and a method for evaluating the contribution to data generation for clear reward distribution.

[0228] Below, a data trading method based on the aforementioned blockchain network and a method for evaluating the contribution to data generation for the distribution of clear rewards are explained in detail.

[0230] FIG. 18 is a diagram illustrating the function of an electronic device to provide a data trading environment according to various embodiments.

[0231] FIG. 19 is a diagram illustrating how an electronic device provides a data trading environment according to various embodiments.

[0232] Referring to FIG. 18 (a), the electronic device can obtain a synthetic data set from a plurality of supplier terminals (S1850). A specific method for generating a synthetic data set from a plurality of supplier terminals has been described above, so it will be omitted.

[0233] For example, referring to FIG. 19, a plurality of supplier terminals (e.g., Supplier #1, Supplier #2, and Supplier #3) can generate a plurality of sub-data sets (1915) using a data generation tool. Additionally, an electronic device can obtain a synthetic data set (1910) based on the plurality of sub-data sets (1915) and a reference data set (1911) provided for data generation.

[0234] Additionally, referring again to FIG. 18(a), the electronic device can acquire and store contribution information including the contribution of each of the plurality of supplier terminals based on the synthetic data set (S1860). At this time, the contribution information may include not only the contribution of each of the plurality of supplier terminals (supplier contribution information) but also other information, the details thereof of which are explained in FIG. 24.

[0235] For example, referring to FIG. 19, the electronic device can generate contribution information (1920) that reflects the degree of data contribution of multiple supplier terminals by judging the synthetic data set (1910) according to a predetermined standard.

[0236] Additionally, referring again to FIG. 18(a), when a data transaction is performed, the electronic device may distribute rewards associated with the data transaction to a plurality of supplier terminals based on the contribution information (S1870). In this case, the data transaction may be performed in the manner shown in FIG. 18(a), but is not limited thereto.

[0237] For example, referring to FIG. 19, the electronic device may distribute rewards (1940) obtained based on data transactions (1930) with consumers to each of the plurality of supplier terminals based on the contribution information (1920). In this case, the consumer may refer to a demand terminal that requested the creation of a data set, but is not limited thereto, and may also refer to a user terminal that intends to purchase a created data set. Additionally, the contribution information (1920) may include a contribution rate assigned to each of the plurality of supplier terminals, and the electronic device may distribute the rewards based on the assigned contribution rate. In addition, in this case, the electronic device may record the contribution rate assigned to the original owner of the synthetic data set (1910) (e.g., a server jointly owned by data providers or to which ownership is entrusted) and the data providers in the first block data (1951) of the blockchain network (1950).

[0239] Additionally, referring to Fig. 18(b), the electronic device can obtain a request for a data transaction (S1810). At this time, the request may be received from a user terminal that wishes to purchase data, but is not limited thereto, and the request may be automatically generated when the smart contract conditions are met.

[0240] Additionally, the electronic device may perform a data transaction when a predetermined criterion is met (S1820). At this time, the predetermined criterion may vary depending on the data transaction method; for example, the predetermined criterion may be set based on whether an agreement regarding the data transaction has been generated or whether a pre-set smart contract condition has been met. Additionally, the electronic device may perform a data transaction by mediating data transactions between user terminals or by directly selling data to data demanders.

[0241] For example, referring to FIG. 19, the electronic device may perform a data transaction (1930) by transferring a synthetic data set (1910) to a consumer and receiving a reward (1940) from the consumer when a predetermined criterion is met. At this time, the reward (1940) may include virtual currency transferred from the consumer and / or virtual currency generated from the blockchain network (1950) from the consumer.

[0242] Additionally, referring again to FIG. 18(b), the electronic device can transmit and store information related to the data transaction to a blockchain network (S1830).

[0243] For example, referring to FIG. 19, the electronic device can record transaction information associated with the data transaction (1930) and the status of changes in the owner in the second block data (1952) of the blockchain network (1950).

[0244] Below, the method for trading synthetic data sets using a blockchain network is explained in detail.

[0247] FIG. 20 is a flowchart illustrating a method for an electronic device to trade a synthetic data set obtained from a supplier terminal using a blockchain network according to various embodiments.

[0248] FIG. 21 is a diagram illustrating information stored in a blockchain network as an electronic device trades a synthetic data set obtained from a supplier terminal according to various embodiments.

[0249] Referring to FIG. 20, an electronic device (e.g., a server) may acquire a first sub-data set from a first user terminal and a second sub-data set from a second user terminal (S2010). Additionally, the electronic device may acquire a first composite data set including the first sub-data set and the second sub-data set (S2020). At this time, the first user terminal and the second user terminal may refer to an electronic device assigned the role of a data provider. The operation according to steps S2010 and S2020 may be performed using the same technical features described in the description of FIGS. 7 to 17.

[0250] Additionally, the electronic device may evaluate the first sub-data set and the second sub-data set according to predetermined criteria to assign a first contribution to the first provider terminal and assign a second contribution to the second provider terminal (S2030). A specific method for assigning contributions to the provider terminals is described below (Fig. 22).

[0251] Additionally, the electronic device can obtain first block data based on the first synthetic data set, the first contribution, and the second contribution (S2040). Specifically, the electronic device can generate first block data based on the first synthetic data set, the first contribution, and the second contribution and transmit it to a blockchain network.

[0252] For example, referring to FIG. 21, an electronic device may generate a first block data (2100a) composed of a HEADER (2110a) and a BODY (2120b). Specifically, the electronic device may generate and transmit to a blockchain network the first block data (2100a) comprising a HEADER (2110a) containing hash information of a previous block and current hash information, and a BODY (2110b) containing first transaction information (2130) generated based on the first synthetic data set, the first contribution, and the second contribution. At this time, the first transaction information (2130) may include ownership information (2131) indicating the owner of the first synthetic data set, share information (2133) indicating the share ratio between co-owners when the first synthetic data set is jointly owned, and first reward information (2135) indicating a reward paid upon generating the first synthetic data set.

[0253] At this time, the ownership information (2131) may include information about the current owner of the first synthetic data set and information about the past owner. When the first synthetic data set is created for the first time, at least one user who created the first synthetic data set may be included in the ownership information (2131) as the first owner of the first synthetic data set. Additionally, when the first synthetic data set is traded, the ownership information (2131) may include information about the change in owner regarding the owner changed by the transaction. For example, the ownership information (2131) may be displayed in a manner that includes the user's ID (e.g., "User #1, User #2").

[0254] Additionally, the share information (2133) may be obtained based on the first contribution and the second contribution. Specifically, the electronic device may obtain the share information (2133) by calculating the share ratio of at least one user based on the first contribution and the second contribution, which are identified based on the degree to which at least one user contributed to generating the first synthetic data set. For example, the share information (2131) may be presented in a manner that lists the share ratios of each user (e.g., User #1 60%, User #2 40%). Furthermore, although not illustrated in FIG. 21, the share information (2133) obtained by the electronic device may further include platform contribution information, which indicates the degree to which the platform device contributed to generating the data. Details regarding this are explained below (Figs. 24 to 29).

[0255] Additionally, the first reward information (2135) may include a reward associated with the creation of the first synthetic data set. Specifically, when the first synthetic data set is created by at least one user terminal, the electronic device may create and provide a predetermined reward in response to the creation of the first synthetic data set. At this time, the provided reward may be recorded in the first block data (2100a). Additionally, the provided reward may be distributed to the at least one user based on the share information (2133). For example, the electronic device may distribute a predetermined reward to the first user and the second user (e.g., 60% for the first user, 40% for the second user) based on the first contribution and the second contribution in response to the creation of the first synthetic data set.

[0256] Additionally, referring again to FIG. 20, the electronic device can acquire second block data based on a transaction agreement for the first synthetic data set (S2050). At this time, the transaction agreement can be achieved by approving a transaction request from a third user terminal. At this time, the third user terminal may refer to an electronic device acting as a demander who requested the creation of the first synthetic data set, or an electronic device acting as a consumer who requested a transaction of the created first synthetic data set.

[0257] For example, data generation is requested from a third user terminal, a first synthetic data set is generated in accordance with the request, a purchase proposal including a reward is received from the third user terminal regarding the generated first synthetic data set, and the transaction agreement can be obtained by transmitting approval for the purchase proposal. Additionally, for example, data generation is requested from a third user terminal, a first synthetic data set is generated in accordance with the request, a sales proposal including a reward is transmitted to the third user terminal regarding the generated first synthetic data set, and the transaction agreement can be obtained by receiving approval for the sales proposal from the third user terminal. Additionally, for example, a first synthetic data set is generated, a transaction proposal regarding the generated first synthetic data set is received from the third user terminal, transaction details including a reward are transmitted in response to the transaction proposal, and approval for the transaction details is received.

[0258] For example, referring to FIG. 21, the electronic device may generate and transmit to a blockchain network a second block data (2100b) containing second transaction information (2140) that reflects a transaction consensus for a first synthetic data set. At this time, the second transaction information (2140) may include owner change information (2141) indicating the status of changes in the owner following the transaction and second reward information (2143) including rewards accompanying the transaction. For example, when a transaction for the first synthetic data set is concluded between a first user, a second user, and a third user, the electronic device may generate owner change information (2141) indicating that the owner of the first synthetic data set has changed from the first user and the second user to the second user, transmit the first synthetic data set to a third user terminal, and generate the second reward information (2143) based on a predetermined reward received from the third user terminal. Additionally, the electronic device can generate a second block data (2100b) including the first transaction information (2130) and the second transaction information (2140) and transmit it to a blockchain network.

[0259] Additionally, referring again to FIG. 20, the electronic device can distribute rewards according to the transaction to the first supplier terminal and the second supplier terminal based on the first contribution and the second contribution (S2060). For example, referring to FIG. 21, the electronic device can distribute rewards to the first user and the second user based on share information (2133) stored in the first block data (2100a).

[0261] FIG. 22 is a flowchart illustrating a method in which an electronic device provides a smart contract-based data trading environment according to various embodiments.

[0262] FIG. 23 is a diagram illustrating a method for generating block data accompanying a smart contract-based data transaction using an electronic device, according to various embodiments.

[0263] Referring to FIG. 22, an electronic device (e.g., a server) may acquire a first block of data based on smart contract information that includes contract conditions set based on request information related to data generation (S2210). At this time, the contract conditions may refer to conditions for the smart contract to proceed with a transaction of the first synthetic data set. Additionally, the contract conditions may be identical to the data generation conditions included in the request information. For example, the contract conditions may be set based on the attributes, volume, quality, or similarity to reference data of the synthetic data set to be generated. Specifically, the electronic device may generate a first block of data that includes smart contract information generated based on the contract conditions—wherein, if the contract conditions are met, a transaction for the first synthetic data set is completed—and transmit it to a blockchain network.

[0264] For example, referring to FIG. 23, the electronic device may generate a first block data (2300a) composed of a HEADER (2310a) and a BODY (2320a). The technical features described above may be applied as they are to the information included in the HEADER and BODY of the block data. Additionally, the first block data (2300a) generated by the electronic device may include first transaction information (2330) containing information related to the creation and ownership of the first synthetic data set, and smart contract information (2340) containing information related to a smart contract for the first synthetic data set. Since the description regarding the transaction information has been described above, it will be omitted. At this time, when a smart contract is generated for the first synthetic data set, the electronic device may generate the smart contract information (2340) including contract conditions (2341) related to the smart contract and party information (2342) of the smart contract. Specifically, the contract conditions (2341) may reflect conditions for a transaction to proceed according to a smart contract for the first synthetic data set, and the party information (2342) may reflect both parties to the transaction according to the smart contract. Additionally, the contract conditions (2341) may include a number of constraints to which the first synthetic data set must correspond (e.g., attributes, volume, quality, or similarity with reference data of the synthetic data set).

[0265] Additionally, referring again to FIG. 22, the electronic device can obtain a first synthetic data set including at least one sub-data set generated from at least one supplier terminal (S2220). Additionally, the electronic device can determine whether the contract conditions are met based on the first synthetic data set and the request information (S2230). Specifically, the electronic device can determine that the contract conditions are met if the data generation conditions included in the request information are met. For example, referring to FIG. 23, it can be determined whether the constraints required by the contract conditions (2341) included in the first synthetic data first block data (2300a) are met.

[0266] Additionally, when the above contract conditions are met, the electronic device may transmit the first synthetic data set to the demander terminal by proceeding with a transaction according to the smart contract (S2240). Specifically, the electronic device may be configured so that a data transaction according to the smart contract is executed as soon as the first synthetic data set meets the above contract conditions. For example, the electronic device may be configured so that the first synthetic data set is transmitted to the demander terminal as soon as the above contract conditions are met, and a reward is transmitted from the demander terminal in response. In this case, the reward transmitted from the demander terminal may be transmitted to the electronic device or the supplier terminal. Additionally, if the first synthetic data set is generated by a plurality of supplier terminals, the reward may be distributed based on the contribution of the plurality of supplier terminals, but is not limited thereto.

[0267] Additionally, the electronic device may acquire second block data based on smart transaction information related to a transaction according to the smart contract (S2250). Specifically, as contract conditions are met, a smart transaction may be executed according to the smart contract, and the electronic device may generate second block data including smart transaction information related to the smart transaction. At this time, the smart transaction information may include information related to the smart contract and the smart transaction.

[0268] For example, referring to FIG. 23, the electronic device may generate a second block data (2300b) composed of a HEADER (2310b) and a BODY (2320b). Specifically, the electronic device may include first transaction information (2330) and smart transaction information (2150) that includes information related to smart contracts and smart transactions. That is, the smart transaction information (2150) may include not only the smart contract information (2340) included in the first block data (2300a), but also information related to smart transactions that occur according to the execution of the smart contract. For example, the electronic device may include, but is not limited to, contract conditions (2141) included in the smart contract information (2340), a smart transaction history (2151) that records the smart transaction by reflecting party information (2342) included in the smart contract information (2340), and owner change information (2152) resulting from the smart transaction.

[0270] Below, we describe the relationship between the time of creation of the synthetic dataset and the time of creation of the block data corresponding to the synthetic dataset.

[0271] An electronic device according to one embodiment may generate block data corresponding to a synthetic data set at the time when the generation of a synthetic data set is completed by at least one user terminal. In this case, the synthetic data set may be traded according to the method according to FIGS. 20 and FIGS. 21, but is not limited thereto.

[0272] In addition, an electronic device according to one embodiment may generate block data corresponding to the synthetic data set at a predetermined time before the generation of the synthetic data set is completed. That is, the electronic device may be configured so that the synthetic data set can be modified even after block data corresponding to the synthetic data set has been generated.

[0273] For example, an electronic device may generate block data corresponding to a synthetic data set at the time it starts generating the synthetic data set. In this case, the synthetic data set may include only a reference data set, and block data corresponding to the synthetic data set may be generated at a time prior to the determination of a provider terminal for generating the synthetic data set. In addition, in this case, the initial owner reflected in the ownership information of the block data may be a server (administrator terminal) that controls the generation of the synthetic data set.

[0274] As another example, the electronic device may generate block data corresponding to the synthetic data set at any point in time when the generation of the synthetic data set is in progress by a plurality of supplier terminals. In this case, the initial owner reflected in the ownership information of the block data may be at least one user terminal participating in the generation of the synthetic data set. The owner may be a shared entity among at least one user terminal participating in the generation of the synthetic data set. In addition,

[0275] As described above, if block data corresponding to the synthetic data set is generated at a predetermined time before the generation of the synthetic data set is completed, the synthetic data set may be traded according to the smart contract of FIGS. 22 and FIGS. 23, but is not limited thereto. In this case, the electronic device may update the synthetic data set according to the data generation of at least one supplier terminal, and may be configured to execute a smart transaction according to the smart contract if the synthetic data set updated at a specific time meets the contract conditions.

[0277] Below, we describe a method for obtaining contribution information for a dataset generated by multiple parties.

[0278] FIG. 24 is a diagram illustrating a general method for an electronic device to acquire contribution information according to various embodiments.

[0279] Referring to FIG. 24, the electronic device can generate contribution information (2400) based on supplier contribution information (2410) and platform contribution information (2420).

[0280] At this time, the supplier contribution information (2410) may be information reflecting the contribution of a data supplier who directly performed the creation of the data set. Specifically, the supplier contribution information (2410) may include quality information (2411) reflecting the contribution related to the quality of the generated data, difficulty information (2413) reflecting the contribution related to the difficulty of generating the data, and volume information (2415) reflecting the contribution related to the amount of data generated.

[0281] Additionally, the platform contribution information (2420) may be information reflecting the contribution of the platform that provided the environment for generating the data set. Specifically, the platform contribution information (2420) may include reference contribution information (2421) associated with the contribution of reference data provided as a reference for the synthetic data to be generated, generation tool contribution information (2423) associated with the contribution of the data generation tool used to generate the synthetic data, and other tool contribution information (2425) associated with the contribution of other tools such as a quality evaluation model or a data imaging tool.

[0282] Accordingly, the electronic device can obtain contribution information (2400) that reflects the contribution of the provider who directly generated the data and the contribution of the platform providing the data generation environment.

[0284] FIG. 25 is a diagram illustrating a method for an electronic device to acquire contribution information according to quality information according to various embodiments.

[0285] Referring to FIG. 25, the electronic device can obtain a first sub-data set from a first supplier terminal and a second sub-data set from a second supplier terminal (S2510). At this time, the electronic device can obtain a synthetic data set based on the first sub-data set and the second sub-data set.

[0286] Additionally, the electronic device can obtain first quality information corresponding to the first sub-data set and second quality information corresponding to the second sub-data set by inputting the first sub-data set and the second sub-data set into a pre-stored quality evaluation model (2500) (S2520). Specifically, for a synthetic data set including the first sub-data set and the second sub-data set, the electronic device can obtain quality information using the quality evaluation model (2500) based on the synthetic data set including the first sub-data set and the second sub-data set in order to verify the contribution of the data provider terminal to the generation of the synthetic data set.

[0287] Additionally, the electronic device can obtain contribution information based on the first quality information and the second quality information (S2530). Specifically, the electronic device can generate contribution information by assigning a first contribution to a first supplier and assigning a second contribution to a second supplier based on the first quality information and the second quality information. At this time, the contribution information may include not only the contribution information of the suppliers but also the contribution information of the platform providing the environment for generating the synthetic data set. For example, if the quality of the first sub-data set generated by the first supplier, identified by the first quality information, is better than the quality of the second sub-data set generated by the second supplier, identified by the second quality information, the electronic device can generate contribution information by assigning a higher contribution to the first supplier.

[0288] Additionally, the electronic device may create a data fund based on the contribution information (S2540). At this time, the data fund may be an intangible asset that defines ownership shares for the synthetic data set based on the contribution. Specifically, multiple suppliers may jointly own the synthetic data set according to the share ratio confirmed by the data fund. By establishing the data fund as described above, the electronic device may store ownership information for the synthetic data set, and when a reward is generated as the synthetic data set is created or traded, the reward may be distributed based on the data fund.

[0290] FIG. 26 is a diagram illustrating a method for an electronic device to generate quality information for a generated data set according to various embodiments.

[0291] Referring to FIG. 26, in order to obtain quality information in step S2520 of FIG. 25, the electronic device may calculate at least one quality indicator based on at least one quality criterion stored in advance (e.g., at least one of a first quality criterion defined based on distance relationships or a second quality criterion defined based on the degree of compliance with generation conditions) (S2521).

[0292] Specifically, the electronic device can verify the distance relationship between data included in the data set and obtain a quality indicator related to distance based on the verified distance relationship. For example, the electronic device can verify at least some of the density, bias, or uniformity of distribution of the data set by calculating the distance between data included in the data set. Additionally, the electronic device can calculate a quality indicator related to distance of the data set based on the density, bias, or uniformity of distribution of the data set. As a specific example, the electronic device can determine that the quality indicator related to distance is higher when the density of the data set is constant, the bias is low, and the distribution is uniform. Additionally, for example, the electronic device can calculate a quality indicator related to condition compliance based on the degree to which the generated data set meets the requested generation conditions. Specifically, the electronic device can determine that the quality indicator related to condition compliance is higher when the data set meets the generation conditions.

[0293] Additionally, the electronic device can generate quality information based on at least one quality indicator (S2523).

[0294] In this way, the electronic device can calculate the contribution of each supplier based on quality by evaluating the quality of data generated by each supplier, and accordingly, obtain supplier contribution information.

[0295] In addition, the electronic device may obtain supplier contribution information by further considering difficulty information and volume information, as illustrated in FIG. 24, but is not limited thereto. Specifically, the electronic device may assign a high contribution to a supplier who has generated high-difficulty data, and may assign a high contribution to a supplier who has generated a large volume of data.

[0297] FIG. 27 is a flowchart illustrating a method for an electronic device to acquire platform contribution information according to various embodiments.

[0298] Referring to FIG. 27, the electronic device can obtain a synthetic data set from at least one supplier terminal (S2710).

[0299] Additionally, the electronic device can verify supplier contribution information based on the synthetic data set (S2720). At this time, the specific method for verifying the supplier contribution information may be applied as is the method disclosed in FIG. 25 and FIG. 26.

[0300] Additionally, the electronic device can verify reference contribution information based on a first contribution criterion (S2730). For example, the first contribution criterion may be set based on at least one of the proportion of the reference data set in the synthetic data set, the amount of the reference data set, the quality of the reference data set, or the similarity of the synthetic data set to the reference data set. Specifically, the electronic device may assign a higher contribution to the reference data set reflected in the reference contribution information as the proportion of the reference data set in the synthetic data set increases, as the amount of the reference data set increases, as the quality of the reference data set improves, and as the similarity between the reference data set and the synthetic data set increases.

[0301] Additionally, the electronic device can verify the generative tool contribution information based on a second contribution criterion (S2740). For example, the second contribution criterion may be set based on at least one of whether a generative tool is used, the attributes of the generative tool used (e.g., the quality of the generative tool, such as whether it is a high-quality generative model or a generative tool with high construction costs), the frequency of use of the generative tool, or the number of generative tools used. Specifically, the electronic device may use high-quality generative tools at a high frequency to generate synthetic data sets, and may assign a higher contribution of the generative tool reflected in the generative tool contribution information as various generative tools are used.

[0302] Additionally, the electronic device can verify other tool contribution information based on a third contribution criterion (S2750). For example, the third contribution criterion may be applied when other tools, such as data imaging tools or quality evaluation tools, are used. Specifically, when other tools, such as data imaging tools or quality evaluation tools, are used in the process of generating a synthetic data set, a certain contribution resulting from the use of said other tools may be assigned as other tool information.

[0303] Additionally, the electronic device can obtain platform contribution information based on reference contribution information, creation tool contribution information, and other tool contribution information (S2760). Additionally, the electronic device can obtain contribution information based on the supplier contribution information and platform contribution information (S2770).

[0304] At this time, the electronic device can generate platform contribution information for each of at least one supplier terminal. Specifically, the electronic device can calculate the platform contribution per supplier based on the extent to which the at least one supplier terminal utilizes the platform environment.

[0305] Additionally, the electronic device can distribute rewards by taking into account platform contribution information. Specifically, the electronic device can distribute rewards generated from the creation or transaction of synthetic data based on supplier contribution information, and can determine the rewards to be finally distributed to each supplier based on platform contribution information. For example, the electronic device can determine the final reward by deducting a platform fee calculated based on platform contribution information from the rewards distributed to the suppliers. As a specific example, the electronic device may first distribute a first reward to a first supplier, and then provide a second reward to the first supplier by deducting the platform fee from the first reward, taking into account the platform contribution of the first supplier.

[0307] FIG. 28 is a diagram illustrating a method for improving data of an electronic device providing a platform according to various embodiments.

[0308] FIG. 29 is a diagram illustrating a method for obtaining platform contribution information as an electronic device providing a platform improves data according to various embodiments. Reference numerals 2910, 2930, and 2940 in FIG. 29 are illustrated as data images by example, but may represent actual data.

[0309] Referring to FIG. 28, an electronic device according to one embodiment may obtain a reference data set based on request information for data generation (S2810). Additionally, the electronic device may obtain a first synthetic data set generated using a first data generation tool (S2820). For example, referring to FIG. 29, a first electronic device (2900a) may obtain a reference data set (2910) including a visual display (2920) representing request information. For example, the first electronic device (2900a) may be a data provider terminal that receives the reference data set reflecting the request information. In this case, the provider terminal may generate data based on the reference data set (2910) according to the request information. Accordingly, a first synthetic data set (2930) may be obtained based on the reference data set (2910).

[0310] Additionally, referring again to FIG. 28, the electronic device can check whether the first synthetic data set satisfies a first predetermined condition (S2830). At this time, the first predetermined condition may be set based on the first synthetic data set and the request information. Specifically, the electronic device may determine that the first predetermined condition is satisfied if the first synthetic data set achieves a generation condition included in the request information above a predetermined threshold.

[0311] For example, referring to FIG. 29, the first predetermined condition may be set based on the amount of data corresponding to the generation condition among the synthetic data generated by the provider. Additionally, the first predetermined condition may be set based on the ratio of data generated within the internal area of ​​the visual display (2920) reflecting the request information. For example, the second electronic device (2900b) may determine that the first predetermined condition is satisfied if the amount of data corresponding to the generation condition or the ratio of data corresponding to the request information in the first synthetic data set is greater than or equal to a threshold. For example, the second electronic device (2900b) may be a server that determines whether the condition of the synthetic data set is met.

[0312] If the first synthetic data set does not satisfy the first predetermined condition, the operation of step S2820 can be performed again to generate a synthetic data set.

[0313] Additionally, referring again to FIG. 28, the electronic device can improve the first synthetic data set by processing the first synthetic data set based on the first algorithm when the first synthetic data set satisfies the first predetermined condition (S2840). Here, improvement of data means modifying or deleting a part of the data included in the data set or creating additional data.

[0314] Specifically, the first algorithm is an algorithm for improving the first synthetic data set, and the specific method of operation of the electronic device (processor) is as follows.

[0315] The electronic device can generate a first set of data points by mapping the first synthetic data set to a first embedding space. At this time, the first set of data points may reflect the distribution of the first synthetic data set. Specifically, each data included in the synthetic data set may correspond to each data point included in the first set of data points. Additionally, the electronic device can obtain a first data image corresponding to the first synthetic data set by displaying the first set of data points in an imaging space.

[0316] Additionally, after generating the first data point set, the electronic device may obtain a first improved data point set by adjusting the attributes of at least some of the data points among the plurality of data points included in the first data point set. At this time, the attributes of the data points may include, but are not limited to, the number of data points, the position of the data points in the embedding space, and the distance between data points. For example, the electronic device may obtain the first improved data point set by adjusting the position in the first embedding space of at least some of the plurality of data points included in the first data point set. Additionally, for example, the electronic device may obtain the first improved data point set by adding a predetermined data point to the first data point set.

[0317] Additionally, after generating the first improved data point set, the electronic device may obtain a first improved synthetic data set based on the first improved data point set. Specifically, the electronic device may generate the first improved synthetic data set using a pre-stored generative model. For example, the electronic device may generate the first improved synthetic data set by decoding the first improved data point set defined in the first embedding space into an input space, but is not limited thereto. As the attributes of the data in the embedding space are adjusted, the first improved synthetic data set may have different attributes from the first synthetic data set (e.g., data of higher quality or data of higher volume, etc.).

[0318] For example, referring to FIG. 29, if the first synthetic data set (2930) satisfies a first predetermined condition, the second electronic device (2900b) can generate a first improved synthetic data set (2940) based on the first synthetic data set.

[0319] Additionally, if the second electronic device is a platform providing device (e.g., a server or an administrator terminal), the second electronic device may obtain platform contribution information. Specifically, the second electronic device may obtain the platform contribution information based on the first synthetic data set (2930) and the first improved synthetic data set (2940). Specifically, the second electronic device may obtain the platform contribution information by reflecting the contribution of the reference data set (2910) and the degree of data improvement according to the first algorithm used to generate the first improved synthetic data set (2940). For example, the second electronic device (2900b) may transmit the final reward, excluding the fee for the improvement from the reward generated by the creation or transaction of the first improved synthetic data set, to the first electronic device (2900a).

[0321] [Data Utilization Environment]

[0322] An electronic device according to various embodiments of the present disclosure can provide a data utilization environment that can utilize various data sets (e.g., synthetic data sets).

[0323] Hereinafter, various embodiments of utilizing data sets generated or traded through the platform environment in the platform environment are described.

[0324] FIG. 30 is a diagram illustrating a method in which an electronic device provides feedback on data generation using a reference model according to various embodiments.

[0325] Referring to FIG. 30, a platform providing device (e.g., a server) can acquire a synthetic data set (3001) (S3010). At this time, the server can receive the synthetic data set (3001) generated from at least one provider terminal.

[0326] Additionally, the server can train a reference model (3005) based on the synthetic data set (3001) (S3020). At this time, the reference model (3005) may be an AI model (e.g., a deep learning model, an NN model, etc.) that is stored in the server in advance. For example, the server can train a waste plastic classification model based on a synthetic data set of waste plastic obtained from at least one supplier terminal.

[0327] Additionally, the server can evaluate the learned reference model (3005) and obtain the results of the evaluation (S3030). Specifically, the server can evaluate the learned reference model to determine whether the learned reference model (3005) works well in actual situations. For example, the server can obtain an evaluation data set (3003) from the first electronic device and obtain results by inputting the evaluation data set (3003) into the learned reference model (3005), and can evaluate the performance of the learned reference model (3005) based on the results. At this time, the evaluation data set (3003) may be data collected from the first electronic device. For example, the first electronic device may obtain the evaluation data set (3003) by capturing an image (e.g., an image of waste plastic) of an object in the same domain as the domain of the synthetic data set (3001). Additionally, for example, the first electronic device may obtain the evaluation data set (3003) by arbitrarily generating data of an object in the same domain as the domain of the synthetic data set (3001).

[0328] Additionally, if the result output for the evaluation data set (3003) through the reference model (3005) does not satisfy a predetermined standard, the server may transmit feedback regarding the result to the at least one supplier terminal (S3040). For example, the server may transmit failure data to the at least one supplier terminal, request the generation of additional data to overcome the failure case, or request the improvement of the generated data, but is not limited thereto. As a specific example, the server may transmit the evaluation data set (3003) as failure data to the at least one supplier terminal and request the additional generation of a synthetic data set associated with the evaluation data set (3003). For example, the server may transmit a message requesting the generation of data similar to the evaluation data set (3003).

[0329] Additionally, the feedback may include a range of generation parameters for generating data associated with the evaluation data set—the generation parameters being used by the data generation tool to generate synthetic data. Specifically, the server may provide a range of generation parameters that must be applied to the data generation tool to generate data similar to the evaluation data set.

[0330] In the training and evaluation process of reference models using training data, a problem may arise where unsuitable results are produced depending on the surrounding environment in which the data was acquired. A typical example is when training image data in the "autonomous driving" domain; an autonomous driving solution trained with image data acquired in Country A may not produce suitable results when input with image data acquired in Country B.

[0331] To address this, an electronic device according to one embodiment of the present disclosure may provide feedback for domain adaptation in the learning of a reference model. Specifically, the electronic device may distribute a reference model learned based on a data set acquired in a first environment to a device used in a second environment. In this case, the electronic device may perform an operation to adapt the domain of the data acquired in the first environment to the domain of the data acquired in the second environment. For example, the electronic device may generate data acquired in the second environment based on data acquired in the first environment, but is not limited thereto. As a specific example, the electronic device may transmit feedback requesting the generation of synthetic data associated with the second environment to at least one data provider terminal.

[0332] In addition, at least one provider terminal can generate a synthetic data set based on the received feedback.

[0334] FIG. 31 is a diagram illustrating how an electronic device learns and demonstrates a waste plastic classification model according to various embodiments.

[0335] Referring to FIG. 31, the electronic device can learn a reference model (3103) based on a generated first synthetic data set (3101) (S3110). At this time, the first synthetic data set (3101) may be an image data set of the waste plastic domain, but is not limited thereto. Additionally, the reference model (3103) may be an AI model for classifying waste plastic according to a predetermined standard, but is not limited thereto. For example, the data related to waste plastic input into the reference model (3103) may be classified into at least one class based on criteria such as contamination level, degree of wrinkling, and whether a label is attached. Additionally, for example, the reference model (3103) may be a model for determining recyclable waste plastic among waste plastics. Specifically, the electronic device may be configured to receive a specific plastic image using the reference model and determine whether the plastic represented by the plastic image is recyclable plastic.

[0336] Additionally, the electronic device can deploy the learned reference model (3103) to a platform environment (S3120). At this time, 'deployment to a platform environment' may mean providing the reference model to users of the platform environment so that they can demonstrate it.

[0337] In addition, the electronic device may deploy the learned reference model (3103) to a real-world environment, not limited thereto. Specifically, the electronic device may deploy the learned reference model (3103) to a real-world environment where a task is performed using the reference model.

[0338] Additionally, at least one user terminal using the platform environment can demonstrate the distributed reference model (3103) based on actual data (S3130). Specifically, the user terminal using the platform environment can input an evaluation data set (3111) into the reference model (3103) to check whether the solution (3100) built using the reference model (3103) works well.

[0339] In addition, in this case, the electronic device can demonstrate the solution (3100) based on the reference model (3103) through a preset output environment (demonstration environment or actual use environment within the platform, 3105). At this time, the preset output environment may be a demonstration environment for simulation within the platform. In addition, not limited thereto, the preset output environment may be a work environment for actual use.

[0340] Specifically, the electronic device can output the input evaluation data set (3111) to the output environment (3105). Additionally, the electronic device can output the result of the solution (3100) based on the reference model (3103) using at least one output device (3107). For example, if the solution (3100) is a device for determining recyclable waste plastic using the reference model (3103), the at least one output device (3107) can display data determined to be recyclable plastic in the output environment (3105). As a specific example, the at least one output device (3107) may include an AR projector. In this case, the at least one output device (3107) can project a visual effect onto the first plastic data (3112) classified as recyclable plastic. Through this, a user demonstrating or using the solution (3100) can visually confirm the recyclable plastic.

[0341] Additionally, the electronic device can retrain the reference model (3103) by providing feedback on failure cases that occurred during the demonstration of the solution (S3140). Specifically, the electronic device can optimize the reference model by additionally learning failure cases for the reference model being demonstrated in a platform environment. For example, if a device for classifying recyclable plastics fails to identify second plastic data (3113) regarding recyclable plastics, the electronic device can store the second plastic data (3113) as a failure case. Additionally, the electronic device can retrain the reference model (3103) based on the second plastic data (3113), generate additional synthetic data similar to the second plastic data (3113), and retrain the reference model (3103) based on the generated synthetic data.

[0342] In addition, but not limited to this, the electronic device may provide feedback to at least one data provider terminal. Specifically, the electronic device may request the at least one provider terminal to generate additional data associated with a failure case of the reference model.

[0344] Below, an embodiment is described in which an electronic device according to one embodiment generates a synthetic data set related to outlier data and utilizes it to train and demonstrate a model for identifying outlier data.

[0345] Here, outlier data is a concept opposite to normal data.

[0346] Normal data can refer to general data within a specific domain. For example, in the context of plastic domain data, plastic images can be considered normal data, and in the context of financial transaction domain data, remittance records within the normal range can be considered normal data.

[0347] Outlier data refers to data that is not typical, peculiar, abnormal, or inappropriate within a specific domain. For example, in the plastic domain, images that do not contain plastic may be outlier data, while in the financial transaction domain, remittance records that exceed the normal range or financial records requiring a determination of involvement in a crime may be considered outlier data.

[0348] In particular, regarding financial data, it is important to accurately identify outlier data to detect phishing crimes and tax evasion.

[0350] FIG. 32 is a flowchart illustrating a data construction method for an electronic device to learn a model for identifying outlier data according to various embodiments.

[0351] Referring to FIG. 32, the electronic device can check a reference data set (S3210). At this time, the reference data set may be data stored in advance in the electronic device and may be normal data identified from outlier data. That is, the reference data set may be data that can be used as normal data to train a model for identifying outlier data.

[0352] Additionally, the electronic device may obtain data generation conditions set based on the reference data set (S3220). At this time, the data generation conditions may be set based on the relationship with the reference data set. Specifically, the data generation conditions may be set based on similarity with the reference data set based on the distance from the reference data set. For example, the electronic device may obtain data generation conditions where the similarity with the reference data set, which is normal data, is set to be below a threshold, but is not limited thereto.

[0353] Additionally, the electronic device may provide the data generation conditions to at least one data provider terminal, and the data provider terminal may acquire synthetic data according to the data generation conditions. (Not shown)

[0354] Additionally, the electronic device can identify a first association based on the acquired synthetic data and the reference data set (S3230). Specifically, to identify the association between the synthetic data and the reference data set, the electronic device can identify a similarity based on the distance between the synthetic data and the reference data set. For example, the electronic device can identify a first association that reflects the similarity between the synthetic data and the reference data set by calculating the distance between the acquired synthetic data and the reference data set.

[0355] Additionally, the electronic device can check whether the first association relationship meets the data generation condition (S3240).

[0356] Additionally, the electronic device may determine the synthetic data as outlier data if the first association corresponds to the data generation condition (S3250). In addition, in this case, the electronic device may provide a reward to at least one supplier terminal that generated the synthetic data for satisfying the data generation condition.

[0357] In addition, the electronic device can use the outlier data for training the model along with the result of determining that the data is the outlier data (S3260).

[0358] Additionally, if the first association does not meet the data generation condition, the electronic device may determine the synthetic data as normal data (S3270). In this case, the electronic device may provide feedback to at least one supplier terminal that generated the synthetic data set that the data generation condition does not meet.

[0359] In addition, the electronic device can use the synthetic data for training the model along with the result of determining that it is normal data (S3280).

[0360] A model for identifying outlier data learned according to the above-described method may be provided to a data provider as a tool for verifying outlier data, or for the purpose of detecting abnormal data in a specific domain.

[0362] FIG. 33 is a diagram illustrating a method for an electronic device to demonstrate a model for determining outlier data according to various embodiments.

[0363] Referring to FIG. 33, the electronic device can input an evaluation data set (3301) into a first artificial intelligence model (3300) for determining outlier data.

[0364] In this case, according to the result output from the first artificial intelligence model (3300), the electronic device can classify the evaluation data set (3301) as normal data based on a predetermined criterion (S3310).

[0365] Additionally, based on the result output from the first artificial intelligence model (3300), the electronic device may classify the evaluation data set (3301) as outlier data based on predetermined criteria (S3320). At this time, the electronic device may provide the outlier data (S3330).

[0366] For example, a user terminal may input a first data set associated with a financial transaction into an electronic device including the first artificial intelligence model (3300). In this case, if the electronic device determines that the first data set is a normal transaction according to a predetermined standard, it may output a result indicating a normal transaction to the user terminal or not output any result. Additionally, if the electronic device determines that the first data set is an abnormal transaction (e.g., a transaction suspected of crime or a remittance) according to a predetermined standard, it may output the result to the user terminal.

[0368] The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., either alone or in combination. The program instructions recorded on the medium may be those specifically designed and configured for the embodiment, or they may be those known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Examples of program instructions include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. The hardware devices described above may be configured to operate as one or more software modules to perform the operation of the embodiment, and vice versa.

[0369] Although the embodiments have been described above with reference to limited examples and drawings, those skilled in the art can make various modifications and variations from the description above. For example, suitable results can be achieved even if the described techniques are performed in a different order than described, and / or the components of the described system, structure, device, circuit, etc. are combined or assembled in a form different from described, or replaced or substituted by other components or equivalents.

[0370] Therefore, other implementations, other embodiments, and equivalents to the claims also fall within the scope of the claims set forth below.

Claims

Claim 1 An electronic device comprising: a memory; and at least one processor electronically connected to the memory and operating based on at least one instruction stored in the memory; wherein the at least one processor obtains a first generation condition for generating data based on a reference data set previously stored in the memory—wherein, the first generation condition is set such that the similarity with the reference data set is below a predetermined threshold—and obtains a first synthetic data generated from a first user terminal, and identifies a first association between the first synthetic data and the reference data set based on the similarity between the first synthetic data and the reference data set, and, based on a judgment that the identified first association corresponds to the first generation condition, is configured to use the first synthetic data for the training of a target artificial intelligence model—wherein the target artificial intelligence model is trained to identify outlier data from a data set. Claim 2 delete Claim 3 An electronic device according to claim 1, characterized in that the first association relationship is obtained by calculating the distance between the first synthetic data and the reference data set. Claim 4 delete Claim 5 delete Claim 6 In claim 1, the at least one processor is an electronic device further configured to provide a reward for the generation of the outlier data to the first user terminal. Claim 7 An electronic device according to claim 1, wherein the at least one processor further comprises acquiring second synthetic data generated from a second user terminal, identifying a second association between the second synthetic data and the reference data set based on similarity between the second synthetic data and the reference data set, and determining the second synthetic data as normal data based on a judgment that the identified second association does not correspond to the first generation condition. Claim 8 In claim 7, the at least one processor is an electronic device further configured to provide feedback to the second user terminal that the generation condition is not met. Claim 9 In claim 7, the electronic device further configured such that the at least one processor inputs the second synthetic data determined as the normal data into a pre-stored artificial intelligence model. Claim 10 A method comprising: a step of obtaining, by at least one processor included in an electronic device, a first generation condition for generating data based on a reference data set stored in memory, wherein the first generation condition is set such that the similarity with the reference data set is below a predetermined threshold; a step of obtaining first synthetic data generated from a first user terminal; a step of identifying a first association between the first synthetic data and the reference data set based on the similarity between the first synthetic data and the reference data set; and a step of determining the first synthetic data as outlier data based on the judgment that the identified first association corresponds to the first generation condition; and wherein the first synthetic data is used for training a target artificial intelligence model—the target artificial intelligence model is trained to identify outlier data from a data set. Claim 11 delete Claim 12 A method according to claim 10, characterized in that the first association relationship is obtained by calculating the distance between the first synthetic data and the reference data set. Claim 13 delete Claim 14 delete Claim 15 A method further comprising the step of providing a reward for the generation of the outlier data to the first user terminal in claim 10. Claim 16 A method according to claim 10 further comprising: a step of obtaining second synthetic data generated from a second user terminal; a step of identifying a second association between the second synthetic data and the reference data set based on similarity between the second synthetic data and the reference data set; and a step of determining the second synthetic data as normal data based on a judgment that the identified second association does not correspond to the first generation condition. Claim 17 A method further comprising, in claim 16, the step of providing feedback to the second user terminal that the generation conditions are not met. Claim 18 A method further comprising, in claim 16, the step of inputting the second synthetic data determined as the normal data into a pre-stored artificial intelligence model. Claim 19 A computer-readable recording medium on which a program for performing the method of paragraph 10 is recorded.

Citation Information

Patent Citations

  • Methods and apparatus for data analysis

    KR1020060026072A

  • System for Detecting Abnormal Control Data

    KR1020200137832A

  • System for Detecting Abnormal Control Data

    KR1020220043319A