Artificial intelligence model-based method and apparatus for providing artificial intelligence modularization service

Abstract

The present invention relates to a method by which an electronic device provides an artificial intelligence modularization service on the basis of an artificial intelligence model. The method comprises generating the artificial intelligence modularization service by using a plurality of artificial intelligence models, that is, an input index evaluation artificial intelligence model and an input data candidate group generation artificial intelligence model.

Description

Method and apparatus for providing AI model-based AI modularization services

[0001] The present invention relates to a method and apparatus for providing an artificial intelligence model-based artificial intelligence modularization service.

[0002] Artificial Intelligence (AI) technology is being actively utilized in various fields. In particular, diverse AI models are being researched and developed to effectively achieve user-desired objectives by utilizing various data obtainable from users.

[0003] In particular, recently, the development of no-code AI solutions, which are tools that help non-experts without coding knowledge easily develop AI-based applications, is actively underway.

[0004] The objective of the present invention is to provide a method and system that recommends prompts based on the level of prompt creation and suggests the development direction of AI services when developing AI services, thereby enabling even non-experts to easily introduce AI models into services.

[0005] The problems of the present invention are not limited to those mentioned above, and other unmentioned problems will be clearly understood by those skilled in the art from the description below.

[0006] According to one embodiment of the present disclosure, in order to solve the problem described above, a method is provided in which an electronic device provides an artificial intelligence modularization service based on an artificial intelligence model. The method comprises the steps of: obtaining a service development request prompt; obtaining service data; determining a first output data and a first input data from the service development request prompt; obtaining an input indicator from the input indicator evaluation artificial intelligence model by inputting the service data, the first output data, and the first input data into an input indicator evaluation artificial intelligence model that outputs an input indicator according to the service data, the first output data, and the first input data when the service data, the first output data, and the first input data are input; and outputting an input data candidate group from the input data candidate group generating artificial intelligence model by inputting the input indicator, the second input data, the third input data, and the fourth input data into an input data candidate group generating artificial intelligence model that outputs an input data candidate group according to the input indicator, the second input data, the third input data, and the fourth input data when the input indicator, the second input data, the third input data, and the fourth input data are input. and the step of generating an artificial intelligence modularization service based on the above input data candidate group; is included.

[0007] According to the features of the present disclosure, the input indicator evaluation artificial intelligence model may be trained to acquire second input data based on the first output data and the service data, and to output the input indicator to any one of stages 1 to 4 based on the first input data and the second input data.

[0008] According to the features of the present disclosure, the artificial intelligence model for generating input data candidate groups can be trained to acquire the third input data, acquire the fourth input data, and output input data candidate groups according to the input indicators.

[0009] According to the features of the present disclosure, the step of generating an artificial intelligence modularization service based on the input data candidate group may include: determining the selected input data candidate group as the fifth input data based on the input of a service creator selecting one or more of the input data candidate groups; and generating an artificial intelligence module layout diagram for generating the artificial intelligence modularization service based on the fifth input data and the first output data.

[0010] According to another embodiment of the present disclosure, in order to solve the problem described above, an electronic device that provides an artificial intelligence modularization service based on an artificial intelligence model may be provided. The electronic device includes a memory that stores one or more instructions and at least one processor that executes the one or more instructions, and the processor, by executing one or more instructions, obtains a service development request prompt; obtains service data; determines a first output data and a first input data from the service development request prompt; and obtains an input indicator from an input indicator evaluation artificial intelligence model by inputting the service data, the first output data, and the first input data to an input indicator evaluation artificial intelligence model that outputs an input indicator according to the service data, the first output data, and the first input data when the service data, the first output data, and the first input data are input; When input indicators, second input data, third input data, and fourth input data are input, an input data candidate group is output from an input data candidate group generating AI model that outputs an input data candidate group based on the input indicators, second input data, third input data, and fourth input data; and an AI modularization service can be generated based on the input data candidate group.

[0011] According to the features of the present disclosure, the input indicator evaluation artificial intelligence model can be trained based on service data training data, first input training data, first output training data, and input indicator training data.

[0012] According to the features of the present disclosure, the artificial intelligence model for generating input data candidate groups can be trained based on input indicator training data, second input training data, third input training data, fourth input training data, and input data candidate group training data.

[0013] Specific details of other embodiments are included in the specific details and drawings for carrying out the invention.

[0014] As explained above, according to the present invention, even non-experts can generate a service function diagram that aligns with the service planning intention, and there is an advantage in that AI models can be easily placed on the service function diagram to conceive a service including an artificial intelligence model.

[0015] Furthermore, according to the present invention, the service creator's understanding of the service is classified as an input indicator based on the degree of creation of a prompt for creating a service, and by presenting a customized service creation direction based on the input indicator and providing an opportunity to redefine the prompt, it is possible to assist the service creator in planning the AI ​​service.

[0016] FIG. 1 is a diagram illustrating a process of providing an artificial intelligence modularization service based on an artificial intelligence model according to one embodiment.

[0017] FIG. 2 is a flowchart illustrating a process of providing an artificial intelligence modularization service based on an artificial intelligence model according to one embodiment.

[0018] FIG. 3 is a flowchart illustrating a specific process of outputting input indicators using an input indicator evaluation AI model in a method of providing an AI modularization service based on an AI model according to one embodiment.

[0019] FIG. 4 illustrates an embodiment in which an electronic device (1000) obtains a prompt (b1) or a prompt (b2).

[0020] FIG. 5 illustrates an example of the case where the electronic device (1000) obtains a prompt (b3).

[0021] FIG. 6 illustrates an example of the case where the electronic device (1000) obtains a prompt (b4).

[0022] FIG. 7 is a flowchart illustrating a specific process of outputting an input data candidate group using an input data candidate group generation AI model in a method of providing an AI modularization service based on an AI model according to one embodiment.

[0023] FIG. 8 is a diagram illustrating the learning process of an input indicator evaluation artificial intelligence model (2000) and an input data candidate group generation artificial intelligence model (3000) according to one embodiment.

[0024] FIGS. 9 and 10 are block diagrams of an electronic device according to one embodiment.

[0025] FIG. 11 is a block diagram of a server according to one embodiment.

[0026] According to one embodiment of the present disclosure, in order to solve the problem described above, a method is provided in which an electronic device provides an artificial intelligence modularization service based on an artificial intelligence model. The method comprises the steps of: obtaining a service development request prompt; obtaining service data; determining a first output data and a first input data from the service development request prompt; obtaining an input indicator from the input indicator evaluation artificial intelligence model by inputting the service data, the first output data, and the first input data into an input indicator evaluation artificial intelligence model that outputs an input indicator according to the service data, the first output data, and the first input data when the service data, the first output data, and the first input data are input; and outputting an input data candidate group from the input data candidate group generating artificial intelligence model by inputting the input indicator, the second input data, the third input data, and the fourth input data into an input data candidate group generating artificial intelligence model that outputs an input data candidate group according to the input indicator, the second input data, the third input data, and the fourth input data when the input indicator, the second input data, the third input data, and the fourth input data are input. and the step of generating an artificial intelligence modularization service based on the above input data candidate group; is included.

[0027] According to the features of the present disclosure, the step of generating an artificial intelligence modularization service based on the input data candidate group may include: determining the selected input data candidate group as the fifth input data based on the input of a service creator selecting one or more of the input data candidate groups; and generating an artificial intelligence module layout diagram for generating the artificial intelligence modularization service based on the fifth input data and the first output data.

[0028] The advantages and features of the present invention and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below but may be implemented in various different forms. These embodiments are provided merely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention, and the present invention is defined only by the scope of the claims. In connection with the description of the drawings, similar reference numerals may be used for similar components.

[0029] In this document, expressions such as "have," "can have," "include," or "can include" refer to the existence of the relevant feature (e.g., numerical values, functions, actions, or components, etc.) and do not exclude the existence of additional features.

[0030] In this document, expressions such as “A or B,” “at least one of A or / and B,” or “one or more of A or / and B” may include all possible combinations of items listed together. For example, “A or B,” “at least one of A and B,” or “at least one of A or B” may refer to cases including (1) at least one A, (2) at least one B, or (3) both at least one A and at least one B.

[0031] Expressions such as “first,” “second,” “first,” or “second” used in this document may modify various components regardless of order and / or importance, and are used only to distinguish one component from another and do not limit said components. For example, the first user device and the second user device may represent different user devices regardless of order or importance. For example, without departing from the scope of rights set forth in this document, the first component may be named the second component, and similarly, the second component may be renamed the first component.

[0032] Where it is stated that a certain component (e.g., a first component) is "(operatively or communicatively) coupled with" or "connected to" another component (e.g., a second component), it should be understood that the said certain component may be directly connected to the said other component or connected through another component (e.g., a third component). On the other hand, where it is stated that a certain component (e.g., a first component) is "directly connected" or "directly connected" to another component (e.g., a second component), it may be understood that no other component (e.g., a third component) exists between the said certain component and the said other component.

[0033] As used in this document, the expression “configured to” may be replaced, depending on the context, with, for example, “suitable for,” “having the capacity to,” “designed to,” “adapted to,” “made to,” or “capable of.” The term “configured to” does not necessarily mean only that which is “specifically designed to” in hardware. Instead, in some situations, the expression “device configured to” may mean that the device is “capable of” in conjunction with other devices or components. For example, the phrase “processor configured to perform A, B, and C” may mean a dedicated processor for performing those operations (e.g., an embedded processor), or a generic-purpose processor (e.g., a CPU or application processor) capable of performing those operations by executing one or more software programs stored in a memory device.

[0034] The terms used in this document are used merely to describe specific embodiments and are not intended to limit the scope of other embodiments. Singular expressions may include plural expressions unless the context clearly indicates otherwise. Terms used herein, including technical or scientific terms, may have the same meaning as generally understood by those skilled in the art described in this document. Terms used in this document that are defined in general dictionaries may be interpreted as having the same or similar meaning as they have in the context of the relevant technology, and are not to be interpreted in an ideal or overly formal sense unless explicitly defined in this document. In some cases, even terms defined in this document may not be interpreted to exclude the embodiments of this document.

[0035] Functions related to artificial intelligence according to the present disclosure are operated through a processor and memory. The processor may be composed of one or more processors. In this case, the one or more processors may be general-purpose processors such as CPUs, APs, and DSPs (Digital Signal Processors), graphics-dedicated processors such as GPUs and VPUs (Vision Processing Units), or artificial intelligence-dedicated processors such as NPUs. The one or more processors control the processing of input data according to predefined operation rules or artificial intelligence models stored in memory. Alternatively, if the one or more processors are artificial intelligence-dedicated processors, the artificial intelligence-dedicated processors may be designed with a hardware structure specialized for processing a specific artificial intelligence model.

[0036] The predefined rules of operation or artificial intelligence models are characterized by being created through learning. Here, being created through learning means that a predefined rules of operation or artificial intelligence models configured to perform a desired characteristic (or objective) are created by a basic artificial intelligence model being trained using multiple learning data by a learning algorithm. Such learning may be performed on the device itself where the artificial intelligence according to the present disclosure is executed, or it may be performed through a separate server and / or system. Examples of learning algorithms include supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning, but are not limited to the examples described above.

[0037] An artificial intelligence model can be composed of multiple neural network layers. Each of the multiple neural network layers has multiple nodes and weight values, and performs neural network operations through calculations between the results of previous layers and the multiple weights. The multiple weights possessed by the multiple neural network layers can be optimized based on the learning results of the artificial intelligence model. For example, multiple weights can be updated so that the loss value or cost value obtained by the artificial intelligence model during the learning process is reduced or minimized. Additionally, to minimize the loss value or cost value, multiple weights can be updated in a direction that minimizes the gradient associated with the loss value or cost value. Artificial neural networks may include deep neural networks (DNNs), such as Convolutional Neural Networks (CNNs), Deep Neural Networks (DNNs), Recurrent Neural Networks (RNNs), Restricted Boltzmann Machines (RBMs), Deep Belief Networks (DBNs), Bidirectional Recurrent Deep Neural Networks (BRDNNs), or Deep Q-Networks, but are not limited to the examples mentioned above.

[0038] The features of each of the various embodiments of the present invention may be combined or combined with one another, either partially or wholly, and as will be fully understood by those skilled in the art, various technical interlocking and operation are possible, and each embodiment may be implemented independently of one another or together in an interlocking relationship.

[0039] Hereinafter, various embodiments of the present invention will be described in detail with reference to the attached drawings.

[0040] FIG. 1 is a diagram illustrating a process of providing an artificial intelligence modularization service based on an artificial intelligence model according to one embodiment.

[0041] Referring to FIG. 1, the electronic device (1000) may use a plurality of artificial intelligence models, namely an input indicator evaluation artificial intelligence model (2000) and an input data candidate group generation artificial intelligence model (3000), to provide an artificial intelligence modularization service. Furthermore, the electronic device (1000) may also use a service generation artificial intelligence model (4000).

[0042] In the present disclosure, the term "artificial intelligence modularization service" refers to a service that provides individual processes for determining individual artificial intelligence models or modules constituting the entire artificial intelligence service, such as a process for receiving and acquiring a group of candidate input data, or a service that generates the entire artificial intelligence service targeted by the artificial intelligence service creator through such processes.

[0043] An electronic device (1000) according to one embodiment may be implemented in various forms. For example, the electronic device (1000) described herein may include a mobile terminal, a smartphone, a laptop computer, a tablet PC, an e-book terminal, a digital broadcasting terminal, a PDA (Personal Digital Assistant), but is not limited thereto.

[0044] Meanwhile, the electronic device (1000) can perform operations to model, learn, modify, or update the input indicator evaluation artificial intelligence model (2000), the input data candidate group generation artificial intelligence model (3000), and the service generation artificial intelligence model (4000).

[0045] According to one embodiment, the artificial intelligence models used by the electronic device (1000), such as the input indicator evaluation artificial intelligence model (2000), the input data candidate generation artificial intelligence model (3000), and the service generation artificial intelligence model (4000), are artificial neural network (ANN) models that refer to a computing system based on biological neural networks. Examples of artificial neural network models include a deep neural network (DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), a restricted boltzmann machine (RBM), a deep belief network (DBN), a bidirectional recurrent deep neural network (BRDNN), and deep Q-networks.

[0046] In the following, the input indicator evaluation artificial intelligence model (2000), the input data candidate group generation artificial intelligence model (3000), and the service generation artificial intelligence model (4000) according to the present disclosure will be described as an example of a Deep Neural Network (DNN) model among artificial neural network models, unless otherwise stated.

[0047] FIG. 2 is a flowchart illustrating a process of providing an artificial intelligence modularization service based on an artificial intelligence model according to one embodiment.

[0048] In step S110, the electronic device (1000) can obtain a service development request prompt.

[0049] In the present disclosure, the term "service development request prompt" refers to a command entered by a service creator, which includes a description of the artificial intelligence service the service creator wishes to develop, and is used by an artificial intelligence model to understand the intent of the service creator's input and generate a service that meets that intent. If one wishes to create a service that recommends perfumes, the service development request prompt may be written, for example, as "I want to create a service that recommends perfumes" or "I want to create a service that recommends perfumes suitable for men in their 20s."

[0050] In step S120, the electronic device (1000) can acquire service data.

[0051] In the present disclosure, the term "service data" refers to a set of data in which information regarding an artificial intelligence service is electronically stored. Specifically, service data may include information regarding various artificial intelligence services, such as (i) information regarding artificial intelligence services stored in a service database server, (ii) information regarding artificial intelligence services provided on the web, and (iii) information regarding artificial intelligence services provided by software satisfying an open source license. Service data may include information regarding input data and output data of a previously created artificial intelligence service or model (module).

[0052] In the present disclosure, the term "artificial intelligence service" refers to a service that operates by utilizing an artificial intelligence model or an artificial intelligence module.

[0053] Referring to FIG. 4, the service data may include information data regarding various AI services that have already been created, provided, or distributed, provided for a fee or free of charge, such as, for example, "an AI service that recommends perfumes suitable for the weather" or "an AI service that analyzes the weather and recommends clothing suitable for the weather."

[0054] According to one embodiment, service data is stored in a service database (DB) server, and an electronic device (1000) can obtain service data from the service database (DB) server.

[0055] In step S130, the electronic device (1000) can determine the first output data and the first input data from the service development request prompt.

[0056] In the present disclosure, the term "output data" refers to data associated with an output target to be provided by an artificial intelligence service. According to one embodiment, the output data may be determined by a keyword for the output target. For example, if the service development request prompt is "I want to create a service that recommends perfumes that match an outfit," the electronic device (1000) may determine the output target to be finally provided by the artificial intelligence service as "perfume (product name, product number, place of sale, etc.)" and determine the keyword for the output target as "perfume," thereby determining the output data as "perfume."

[0057] The first output data refers to output data (included in the service development request prompt) determined by the electronic device (1000) based on the service development request prompt.

[0058] In the present disclosure, the term "input data" refers to data related to an analysis target to be analyzed in an artificial intelligence service (in order to output output data appropriately or according to the purpose of the service creator). According to one embodiment, the input data may be determined as a keyword for the analysis target. For example, if the service development request prompt is "I want to create a service that recommends perfumes that match the outfit," the electronic device (1000) may determine the analysis target to be analyzed to recommend perfumes as "clothing shape, type of clothing, color of clothing, etc." and determine the keyword for the analysis target as "outfit."

[0059] The first input data refers to input data (included in the service development request prompt) determined by the electronic device (1000) based on the service development request prompt.

[0060] Furthermore, if the electronic device (1000) determines that there is no subject to analysis in the service development request prompt, it may determine the first input data as "not existing." For example, if there is no subject to analysis in the service development request prompt, such as "I want to create a service that recommends perfumes," the electronic device (1000) may determine the first input data as "not existing."

[0061] In step S140, the electronic device (1000) or the input indicator evaluation artificial intelligence model (2000) can obtain an input indicator from the input indicator evaluation artificial intelligence model by inputting the service data, the first output data, and the first input data to the input indicator evaluation artificial intelligence model that outputs an input indicator according to the service data, the first output data, and the first input data when the service data, the first output data, and the first input data are input.

[0062] FIG. 3 is a flowchart illustrating the specific process of the step (S140) of obtaining (outputting) input indicators using an input indicator evaluation AI model (2000) in a method of providing an AI modularization service based on an AI model according to one embodiment.

[0063] In step S141, the electronic device (1000) or the input indicator evaluation artificial intelligence model (2000) can obtain second input data based on first output data and service data.

[0064] In the present disclosure, the term "second input data" refers to data determined based on information associated with an artificial intelligence service included in the service data, and means a set of input data of an artificial intelligence service having output data identical to the first output data among the information associated with the artificial intelligence service.

[0065] Service data includes information regarding various artificial intelligence services, and service data may include matching data (data sets) in which input data and output data are matched for each artificial intelligence service or model. Referring to FIG. 4, service data may include various artificial intelligence service data such as "an artificial intelligence service that recommends perfume suitable for the weather (a1)" and "an artificial intelligence service that recommends perfume by analyzing an image (a2)," and for each artificial intelligence service data, matching data may include information in which input data and output data are matched, such as "weather-perfume (a1)" and "image-perfume (a2)."

[0066] According to one embodiment, an input indicator evaluation artificial intelligence model (2000) may be trained to acquire second input data based on the first output data and the service data, and to output the input indicator to any one of stages 1 to 4 based on the first input data and the second input data.

[0067] Referring to FIG. 4, for example, an input indicator evaluation artificial intelligence model (2000) can obtain "weather, image, job type, retention power," which is a set of input data of services a1 to a4 (i.e., artificial intelligence models or modules) having the same output data as "perfume," which is the first output data among the service data, as second input data.

[0068] In step S142, the electronic device (1000) or the input indicator evaluation artificial intelligence model (2000) can output an input indicator to any one of steps 1 to 4 based on the first input data and the second input data.

[0069] Specifically, the input indicator evaluation artificial intelligence model (2000) can be trained to output an input indicator to one of four levels, depending on whether the first input data and the second input data exist and whether the first input data is included in the second input data.

[0070] Specifically, the input indicator evaluation artificial intelligence model (2000) can be trained to (i) output an input indicator at level 1 when the first input data does not exist and the second input data exists, (ii) output an input indicator at level 2 when both the first input data and the second input data exist and the first input data is included in the second input data, (iii) output an input indicator at level 3 when both the first input data and the second input data exist and the first input data is not included in the second input data, and (iv) output an input indicator at level 4 when the first input data exists but the second input data does not exist.

[0071] Referring to FIG. 5, when the first input data does not exist as in prompt (b3) but the second input data exists, the input indicator evaluation artificial intelligence model (2000) can output the input indicator in one step.

[0072] Referring to FIG. 4, if there is a first input data (weather) as in prompt (b1) and a second input data, and the first input data (weather) is included in the second input data (weather, image, type of workplace, retention power), the input indicator evaluation artificial intelligence model (2000) can output the input indicator in two stages.

[0073] Referring to FIG. 4, if there is a first input data (location) as in prompt (b2) and a second input data, and the first input data (location) is not included in the second input data (weather, image, type of workplace, retention power), the input indicator evaluation artificial intelligence model (2000) can output the input indicator in three stages.

[0074] Referring to FIG. 6, if there is a first input data and a first output data (fragrance) as in prompt (b4), but there is no service (i.e., an artificial intelligence model or module) among the artificial intelligence services (i.e., artificial intelligence models or modules) included in the service data that has the same output data as the first output data (fragrance) and therefore no second input data, the input indicator evaluation artificial intelligence model (2000) can output the input indicator in 4 steps.

[0075] Accordingly, in the method for providing an artificial intelligence modularization service according to the present invention, criteria (input data) for selecting an artificial intelligence model or module can be provided differently so that the artificial intelligence service creator can indirectly confirm their understanding of the artificial intelligence service they wish to develop from a prompt, and accordingly create the artificial intelligence service. In particular, in the method for providing an artificial intelligence modularization service according to the present invention, the artificial intelligence service creator can create an artificial intelligence service by determining individual artificial intelligence models or modules to be deployed in the configuration of the artificial intelligence service using the provided candidate input data.

[0076] In step S150, the electronic device (1000) or the input data candidate group generating artificial intelligence model (3000) can output an input data candidate group from the input data candidate group generating artificial intelligence model (3000) by inputting the input indicator to the input data candidate group generating artificial intelligence model that outputs an input data candidate group according to the input indicator when the input indicator is input.

[0077] FIG. 7 is a flowchart illustrating a specific process of outputting an input data candidate group (S150) using an input data candidate group generation AI model in a method of providing an AI modularization service based on an AI model according to one embodiment.

[0078] In step S151, the electronic device (1000) or the artificial intelligence model (3000) that generates the input data candidate group can acquire the third input data.

[0079] In the present disclosure, the term "third input data" refers to input data associated with the first output data and including information about a trend or trend keyword associated with the first output data.

[0080] According to one embodiment, an artificial intelligence model (3000) for generating input data candidate groups can be trained to determine a trend keyword and to determine (acquire) a third input data based on the trend keyword.

[0081] Specifically, the input data candidate generation artificial intelligence model (3000) searches for first output data from various media capable of collecting data for determining trend keywords, and collects keywords associated with the first output data from the search results, and can determine a predetermined number, for example, the top 10 keywords, among the collected keywords based on exposure frequency, search frequency, etc. Media capable of collecting data for determining trend keywords include, but are not limited to, social media, the web, smartphone applications, etc.

[0082] For example, if the first output data is "perfume," the input data candidate generation AI model (3000) searches for "perfume" on Instagram and Naver Shopping, and if the search results, such as "perfume using personal color" and "selection of perfume based on location," are collected as trend data, the "personal color" and "location" with the highest exposure frequency among the trend data can be determined as keyword data and obtained as the third input data.

[0083] In step S152, the electronic device (1000) or the artificial intelligence model (3000) that generates the input data candidate group can acquire the fourth input data.

[0084] In the present disclosure, the term "fourth input data" refers to data determined based on information related to an artificial intelligence service included in the service data, and means input data excluding the first input data from a set of input data of an artificial intelligence service that includes the first input data as input data while having output data identical to the first output data among the information related to the artificial intelligence service.

[0085] According to one embodiment, an artificial intelligence model (3000) for generating input data candidate groups can be trained to determine (acquire) a fourth input data based on a service development request prompt and service data.

[0086] For example, referring to FIG. 4, when a prompt (b1) is input, the first input data is "weather" and the first output data is "perfume". Among the artificial intelligence services a1 to a6 included in the service data, the artificial intelligence services that have the same output data as "perfume" and include the first input data in the input data are a1 and a4. The electronic device (1000) or the artificial intelligence model (3000) that generates the input data candidate group can obtain "sustainability" as the fourth input data, excluding "weather" which is the first input data from "weather, sustainability," which is the set of input data of a1 and a4.

[0087] That is, the fourth input data can provide the service creator with information regarding whether the AI ​​service being used in a previously created AI service, which includes the first input data and first output data of interest to the AI ​​service creator, additionally utilizes other analysis targets.

[0088] In step S153, the electronic device (1000) or the artificial intelligence model (3000) that generates input data candidate groups can output input data candidate groups according to the input indicators.

[0089] An artificial intelligence model (3000) for generating input data candidate groups can be trained to output the most suitable input data candidate group for an artificial intelligence service provider based on an input indicator and second to fourth input data. At this time, the input data candidate groups can be output with priority.

[0090] The artificial intelligence model (3000) for generating input data candidate groups can be trained to: (i) output an input data candidate group based on second input data when the input indicator is at level 1; (ii) output an input data candidate group based on second input data, third input data, and fourth input data when the input indicator is at level 2; (iii) output an input data candidate group based on second input data and third input data when the input indicator is at level 3; and (iv) output an input data candidate group based on first input data when the input indicator is at level 4.

[0091] According to one embodiment, an artificial intelligence model (3000) for generating input data candidate groups can output an input data candidate group based on a second input data when (i) the input indicator is at level 1.

[0092] When the input indicator is at level 1, that is, when the AI ​​service creator (user) inputs a prompt (b3) in which no first input data exists, the level of understanding of the AI ​​service that the AI ​​service creator wants to develop can be estimated to be low. Accordingly, when the input indicator is at level 1, the AI ​​model (3000) for generating input data candidate groups provides the AI ​​service creator with input data (second input data) related to the analysis target used in the previously created AI service or model (module) that is included in the service data as input data candidate groups, thereby providing the AI ​​service creator with criteria (input data) for selecting an AI model or module so that the AI ​​service creator can create at least a basic and general-purpose AI service for the service related to the first output data.

[0093] According to one embodiment, an artificial intelligence model (3000) for generating input data candidates can output an input data candidate set including a priority determined based on usage indicators and satisfaction indicators.

[0094] Specifically, referring to FIG. 4, the artificial intelligence services (a1 to a6) included in the service data include a usage indicator (E1) that quantifies the frequency of use by service users regarding the service, and a satisfaction indicator (E2) that quantifies the satisfaction of service users regarding the service. When the input data candidate group generation artificial intelligence model (3000) receives a prompt (b1), in the process of outputting the second input data, "weather, image, type of job, and retention power," as the first input data candidate group, the input data of the artificial intelligence service with a higher sum of the usage indicator (E1) and the satisfaction indicator (E2) can be output as an input data candidate group with a higher priority. In FIG. 4, since the sum of the usage indicator (E1) and the satisfaction indicator (E2) is large as a1, a4, a2, and a3, the input data candidate group generation artificial intelligence model (3000) can output an input data candidate group with the priority of the input data candidate group in the order of weather, retention power, image, and type of job (weather has the highest priority and type of job has the lowest priority).

[0095] Accordingly, the creator of an artificial intelligence service can compensate for a low understanding of artificial intelligence services by receiving criteria (input data) for selecting an artificial intelligence model or module capable of creating at least basic and general-purpose services, as well as criteria for selecting input data with higher user usage frequency and satisfaction and an artificial intelligence model (module) that utilizes said input data. According to one embodiment, the artificial intelligence model (3000) for generating input data candidate groups can output an input data candidate group based on the second input data, the third input data, and the fourth input data when (ii) the input indicator is at level 2.

[0096] When the input indicator is at level 2, that is, when a prompt (b1) is entered in which there exists a pre-existing AI service or model (module) in which the service the AI ​​service creator intends to create, the analysis target (input data), and the output target (output target) are all identical, the AI ​​service creator's understanding of the AI ​​service they wish to develop can be estimated to be at an intermediate level. Accordingly, when the input indicator is at level 2, the AI ​​model (3000) for generating input data candidate groups can provide the AI ​​service creator with input data candidate groups in addition to the second input data, including input data related to trend keywords included in the service data (third input data) and data that the AI ​​service creator may have overlooked or can use as an additional analysis target (fourth input data), thereby providing criteria (input data) for selecting an AI model or module so that the AI ​​service creator can create an AI service that is differentiated by reflecting trends and has a supplemented analysis target for the service related to the first output data.

[0097] According to one embodiment, the artificial intelligence model (3000) for generating input data candidate groups can be trained to output the fourth input data with priority over the third input data during the process of outputting the second input data candidate group.

[0098] According to one embodiment, the artificial intelligence model (3000) for generating input data candidate groups can output an input data candidate group based on the second input data and the third input data when the input indicator is at level 3.

[0099] When the input indicator is at level 3, that is, when the AI ​​service creator inputs a prompt (b2) in which there is no existing AI service or model (module) in which the analysis target (input data) and output target (output data) of the service to be created are both identical, it can be assumed that the AI ​​service creator's understanding of the AI ​​service they wish to develop is unclear but not low. Accordingly, when the input indicator is at level 3, the AI ​​model (3000) for generating input data candidate groups can provide the AI ​​service creator with input data (third input data) related to trend keywords included in the service data, even if it is not data that the AI ​​service creator has overlooked or can use as an additional analysis target (fourth input data), in addition to the second input data, as input data candidate groups, thereby providing the AI ​​service creator with criteria (input data) for selecting an AI model or module so that the AI ​​service creator can create an AI service in which the future value of the AI ​​service is maximized by reflecting the trend as much as possible regarding the service related to the first output data.

[0100] According to one embodiment, the artificial intelligence model (3000) for generating input data candidate groups can be trained to output a third input data with priority over the second input data during the process of outputting a second input data candidate group.

[0101] According to one embodiment, the artificial intelligence model (3000) for generating input data candidate groups can output an input data candidate group based on the first input data when (iv) the input indicator is at level 4.

[0102] When the input indicator is at level 4, there is no artificial intelligence model (module) or service that has output data identical to the first output data, and a new artificial intelligence service needs to be developed. In order to search for and develop an artificial intelligence model (module) suitable for the new artificial intelligence service, the input data candidate generation artificial intelligence model (3000) can output an input data candidate based on the first input data. Accordingly, the input data candidate generation artificial intelligence model (3000) can provide the artificial intelligence service creator with criteria (input data) that allow them to select an artificial intelligence model or module in a direction that reflects trends as much as possible, increases the future value of the artificial intelligence service, and effectively analyzes the target of analysis (input data).

[0103] In step S160, the electronic device (1000) can generate an artificial intelligence modularization service based on the input data candidate group.

[0104] Step S160 may include: determining a selected group of input data candidates as the fifth input data based on input from a service creator selecting one or more of the above input data candidates; and generating an artificial intelligence module layout diagram for creating the artificial intelligence modularization service based on the fifth input data and the first output data.

[0105] According to one embodiment, the electronic device (1000) can obtain an artificial intelligence modularization service by determining a service layout based on fifth input data and first output data, and by placing a modularized AI module in the determined service layout.

[0106] Specifically, the electronic device (1000) can generate an artificial intelligence modularization service from the service generating artificial intelligence model (4000) by inputting the fifth input data and the first output data to the service generating artificial intelligence model (4000) which outputs an artificial intelligence modularization service according to the fifth input data and the first output data when the fifth input data and the first output data are input.

[0107] For example, if the fifth input data is "personal color" and the first output data is "perfume," the service-generating AI model can generate an "AI modular service that analyzes personal color and recommends perfumes suitable for the analyzed personal color."

[0108] Specifically, when the service generating artificial intelligence model (4000) receives the fifth input data and the first output data, it determines a service layout based on the fifth input data and the first output data, and can generate an artificial intelligence modularization service by placing a modularized AI module in the determined service layout.

[0109] In the present disclosure, a service layout refers to a workflow in which the functions of a service are classified and arranged to place an AI module in the service. For example, a service layout for an "AI modular service that analyzes personal color and recommends a perfume suitable for the analyzed personal color" may be a workflow in which detailed functions such as "user image input," "analysis of personal color from user image," "selection of perfume based on personal color," and "output of information regarding perfume" are arranged. In this case, the service creation AI model (4000) can create an AI modular service by placing an "AI module capable of receiving user images," an "AI module that analyzes personal color from user images," an "AI module that outputs a perfume suitable for personal color," and an "AI module that outputs information about the output perfume" into the workflow.

[0110] The service generation artificial intelligence model (4000) can be trained to output an artificial intelligence modularization service according to the fifth input data and the first output data when the fifth input data and the first output data are input.

[0111] FIG. 8 is a diagram illustrating the learning process of an input indicator evaluation artificial intelligence model (2000) and an input data candidate group generation artificial intelligence model (3000) according to one embodiment.

[0112] Referring to FIG. 8, an input indicator evaluation artificial intelligence model (2000) can be trained based on service data training data, first input training data, first output training data, and input indicator training data. Specifically, an electronic device (1000) can match the service data training data, the first input training data, the first output training data, and the input indicator training data, and train the input indicator evaluation artificial intelligence model (2000) to output an input indicator based on the matched service data training data, the first input training data, the first output training data, and the input indicator training data. The process of training the input indicator evaluation artificial intelligence model (2000) can be applied in the same way during the process of modifying and updating the input indicator evaluation artificial intelligence model (2000).

[0113] Referring to FIG. 8, an artificial intelligence model (3000) for generating input data candidate groups can be trained based on input indicator training data, second input training data, third input training data, fourth input training data, and input data candidate group training data. Specifically, an electronic device (1000) can match the input indicator training data, second input training data, third input training data, fourth input training data, and input data candidate group training data, and train the artificial intelligence model (3000) for generating input data candidate groups to output different input data candidate groups according to the input indicator based on the matched input indicator training data, second input training data, third input training data, fourth input training data, and input data candidate group training data. The process of training the artificial intelligence model (3000) for generating input data candidate groups can be applied in the same way during the process of modifying and updating the artificial intelligence model (3000) for generating input data candidate groups.

[0114] FIGS. 9 and 10 are block diagrams of an electronic device according to one embodiment.

[0115] Referring to FIG. 9, an electronic device (1000) according to one embodiment may include a processor (510) and a memory (520). However, not all illustrated components are essential. The electronic device (1000) may be implemented with more components than illustrated, or with fewer components. For example, as illustrated in FIG. 6, an electronic device (1000) according to one embodiment may further include a user input unit (610), a communication unit (620), and a display (630).

[0116] The processor (510) controls the overall operation of the electronic device (1000) by executing one or more instructions in memory (520). For example, the processor (510) can control the user input unit (610), communication unit (620), display (630), etc., by executing one or more instructions stored in memory (520). Additionally, the processor (510) can perform the operation and function of the electronic device (1000) described in relation to FIGS. 1 to 8 by executing one or more instructions stored in memory (520).

[0117] The processor (510) may be composed of one or more processors, and the one or more processors may be general-purpose processors such as CPUs, APs, DSPs (Digital Signal Processors), graphics-only processors such as GPUs, VPUs (Vision Processing Units), or artificial intelligence (AI)-only processors such as NPUs. According to one embodiment, when the processor (510) is implemented as a plurality of processors or graphics-only processors or artificial intelligence-only processors such as NPUs, at least some of the plurality of processors or graphics-only processors or artificial intelligence-only processors such as NPUs may be installed in an electronic device (1000) and another electronic device or server (5000) connected to the electronic device (1000).

[0118] According to one embodiment, a processor (510) includes a memory for storing one or more instructions and at least one processor for executing said one or more instructions, and by executing said one or more instructions, obtains a service development request prompt; obtains service data; determines a first output data and a first input data from said service development request prompt; obtains an input indicator from said input indicator evaluation AI model by inputting said service data, first output data, and first input data into an input indicator evaluation AI model that outputs an input indicator according to said service data, first output data, and first input data when said service data, first output data, and first input data are input; and outputs an input data candidate group from said input data candidate group generation AI model by inputting said input indicator, said second input data, said third input data, and fourth input data into an input data candidate group generation AI model that outputs an input data candidate group according to said input indicator, second input data, third input data, and fourth input data when said input indicator, second input data, third input data, and fourth input data are input; And based on the above input data candidate group, an artificial intelligence modularization service can be generated.

[0119] The memory (520) may include one or more instructions for controlling the operation of the electronic device (1000). The memory (520) may include artificial intelligence models used by the electronic device (1000), for example, an input indicator evaluation artificial intelligence model (2000) and an input data candidate generation artificial intelligence model (3000).

[0120] According to one embodiment, the memory (520) may include at least one type of storage medium among, for example, a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (e.g., SD or XD memory, etc.), RAM (Random Access Memory), SRAM (Static Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic memory, a magnetic disk, and an optical disk, but is not limited thereto.

[0121] The user input unit (610) can receive user input for controlling the operation of the electronic device (1000). For example, the user input unit (610) may include a key pad, a dome switch, a touch pad (contact capacitive method, pressure resistive method, infrared sensing method, surface ultrasonic conduction method, integral tension measurement method, piezo effect method, etc.), a jog wheel, a jog switch, etc., but is not limited thereto.

[0122] The communication unit (620) may include one or more communication modules for communication with the server (5000). For example, the communication unit (620) may include at least one of a short-range communication unit or a mobile communication unit.

[0123] The short-range wireless communication unit may include, but is not limited to, a Bluetooth communication unit, a BLE (Bluetooth Low Energy) communication unit, a Near Field Communication unit, a WLAN (Wi-Fi) communication unit, a Zigbee communication unit, an infrared (IrDA, infrared Data Association) communication unit, a WFD (Wi-Fi Direct) communication unit, an UWB (ultra wideband) communication unit, an Ant+ communication unit, etc.

[0124] A mobile communication unit transmits and receives wireless signals with at least one of a base station, an external terminal, and a server on a mobile communication network. Here, the wireless signal may include various forms of data resulting from voice call signals, video call call signals, or the transmission and reception of text / multimedia messages.

[0125] The display (630) can display information processed by the electronic device (1000). For example, the display (630) can display an interface for controlling the electronic device (1000), an interface for displaying the status of the electronic device (1000), etc.

[0126] FIG. 11 is a block diagram of a server according to another embodiment.

[0127] According to one embodiment, the electronic device (1000) can be connected to and communicated with a server (5000) that provides an artificial intelligence modularization service based on an artificial intelligence model.

[0128] The server (5000) may include a communication interface (5100), a database (5200), and a processor (5300). For example, the communication interface (5100) of the server (5000) according to the present disclosure may correspond to a communication unit (620) of an electronic device (1000), the database (5200) of the server (5000) may correspond to a memory (510) of an electronic device (1000), and the processor (5300) of the server (5000) may correspond to a processor (510) of an electronic device (1000). Additionally, the processor (4300) of the server (5000) may perform a method of providing an artificial intelligence modularization service based on an artificial intelligence model described in relation to FIGS. 1 to 8.

[0129] Although embodiments of the present invention have been described in more detail with reference to the attached drawings, the present invention is not necessarily limited to these embodiments and may be modified in various ways within the scope of the technical spirit of the present invention. Accordingly, the embodiments disclosed in the present invention are intended to explain, not limit, the technical spirit of the present invention, and the scope of the technical spirit of the present invention is not limited by these embodiments. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. Various modifications and improvements by those skilled in the art using the basic concept of the present disclosure as defined in the following claims also fall within the scope of the rights of the present disclosure.

Claims

1. In a method for an electronic device to provide an artificial intelligence modularization service based on an artificial intelligence model, Step of obtaining a service development request prompt; Step to acquire service data; A step of determining the first output data and the first input data from the above service development request prompt; A step of obtaining an input indicator from an input indicator evaluation artificial intelligence model by inputting the service data, the first output data, and the first input data into an input indicator evaluation artificial intelligence model that outputs an input indicator according to the service data, the first output data, and the first input data when the service data, the first output data, and the first input data are input; A step of outputting an input data candidate group from an input data candidate group generating AI model by inputting the input indicator, the second input data, the third input data, and the fourth input data into the AI ​​model that outputs an input data candidate group according to the input indicator, the second input data, the third input data, and the fourth input data when the input indicator, the second input data, the third input data, and the fourth input data are input; and A method comprising the step of generating an artificial intelligence modularization service based on the above input data candidate group.

2. In Paragraph 1, The above input indicator evaluation AI model is, Based on the first output data and the service data above, second input data is obtained, and A method trained to output the input indicator to any one of steps 1 to 4 based on the first input data and the second input data.

3. In Paragraph 1, The above AI model for generating input data candidates is, Acquire the above third input data, Acquire the above-mentioned fourth input data, and A method trained to output a candidate set of input data according to the above input indicators.

4. In Paragraph 1, The step of generating an artificial intelligence modularization service based on the above input data candidate group is: A step of determining the selected input data candidate group as the fifth input data based on the input of a service creator selecting one or more of the above input data candidate groups; and A method comprising the step of generating an artificial intelligence module layout diagram for generating the artificial intelligence modularization service based on the fifth input data and the first output data.

5. In an electronic device that provides an artificial intelligence modularization service based on an artificial intelligence model, Memory for storing one or more instructions, and It includes at least one processor that executes one or more of the above instructions, The processor executes one or more instructions, Obtain a service development request prompt; Acquire service data; Determining the first output data and the first input data from the above service development request prompt; When service data, first output data, and first input data are input, an input indicator is obtained from an input indicator evaluation artificial intelligence model by inputting said service data, first output data, and first input data into an input indicator evaluation artificial intelligence model that outputs an input indicator according to said service data, first output data, and first input data; An input data candidate generation AI model that outputs an input data candidate set according to the input indicator, second input data, third input data, and fourth input data when the input indicator, second input data, third input data, and fourth input data are input, thereby outputting an input data candidate set from the input data candidate generation AI model by inputting the input indicator, the second input data, the third input data, and the fourth input data; and An electronic device that generates an artificial intelligence modularization service based on the above input data candidate group.

6. In Paragraph 5, The above input indicator evaluation artificial intelligence model is an electronic device trained based on service data training data, first input training data, first output training data, and input indicator training data.

7. In Paragraph 5, The above-mentioned input data candidate generation artificial intelligence model is an electronic device trained based on input indicator training data, second input training data, third input training data, fourth input training data, and input data candidate training data.

Images