A method for integrating artificial intelligence multi-models for the digitally excluded

The method facilitates easy integration and utilization of multiple AI models by providing a server-terminal interface for model selection, preprocessing, and accuracy judgment, enhancing user experience and model performance.

KR1020260163147APending Publication Date: 2026-11-16YEJIN
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
KR1020250059028
Authority / Receiving Office
KR · KR
Patent Type
Applications
Current Assignee / Owner
Filing Date
2025-05-07
Publication Date
2026-11-16

AI Technical Summary

Technical Problem

Existing AI models are difficult to access and integrate due to training and usage complexities, limiting their widespread utilization and market growth potential.

Method used

A method for integrating multiple AI models through a server-terminal interface, including model list output, selection, preprocessing, execution variable setting, and result output, with accuracy judgment and integration method candidates, enabling users to easily select and combine AI models for improved performance.

Benefits of technology

Enables users to obtain more accurate results by selecting and combining AI models, ensuring accuracy through re-execution if needed, and providing sample data and integration method candidates for easier model selection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure PAT00010_ABST
    Figure PAT00010_ABST
Patent Text Reader

Abstract

The present invention relates to a method for integrating multiple artificial intelligence models, comprising: a model list output step (S100); a selection transmission step (S200); a preprocessing step (S300) for preprocessing data received from the server (10) after the selection transmission step (S200); an execution variable setting step (S400) for setting execution variables of the artificial intelligence model in the server (10) after the preprocessing step (S300); a model execution step (S500) for executing the selected artificial intelligence model based on data input by the user in the server (10) to obtain a result value after the execution variable setting step (S400); and a result output step (S600) for outputting the result value obtained from the server (10) to the user terminal (20) after the model execution step (S500). In the model list output step (S100), the number of artificial intelligence models included in the list of available artificial intelligence models transmitted by the server (10) to the user terminal (20) is at least two, and the user [uses] artificial intelligence through the user terminal (20). The method comprises: a method integration candidate transmission step (S310) in which at least two models are selected, and after the preprocessing step (S300) and before the execution variable setting step (S400), a candidate for an artificial intelligence model integration method is transmitted from the server (10) to the user terminal (20); and after the method integration candidate transmission step (S310) and before the execution variable setting step (S400), if the user selects one of the candidates for an artificial intelligence model integration method, a model integration step (S320) in which at least two artificial intelligence models are integrated using the selected artificial intelligence model integration method.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present invention relates to a method for integrating multiple artificial intelligence models for the digitally marginalized. More specifically, it relates to a method devised to enable users to more easily use various types of artificial intelligence models. Background Technology

[0003] Over the past few years, the artificial intelligence market has experienced explosive growth, and driven by this rapid expansion, numerous companies are jumping into the development of AI technology. To maintain competitiveness, these companies are developing individual AI models in different ways and are opening up some of the resulting products for use in limited spaces and in limited ways. Considering the characteristics of AI, where the combination of various data and AI models can lead to the emergence of diverse services and increased model performance, it is difficult to view such a closed environment as having a positive effect on market growth. Furthermore, while AI models are very useful in themselves, they are difficult to access for many people due to the difficulty of training and using them. The method for integrating and using multiple AI models according to the present invention is designed to enable users to use various types of AI models more easily. Related prior art is presented below.

[0004] Prior art documents 001 to 004 are inventions that have a technical relationship with the present invention, prior art document 001 relates to an artificial intelligence vulnerability diagnosis learning recommendation service system and method, prior art document 002 relates to a method for automatically analyzing fine particles distributed over a large area using artificial intelligence, an electron microscope, and an EDS analyzer, prior art document 003 relates to a game quality assurance method and system using an artificial intelligence agent, and prior art document 004 relates to an artificial intelligence-based brain disease diagnosis method and system using human dynamic characteristic information.

[0005] Prior art documents 001 to 004 are similar to the method of using multiple AI models integrated according to the present invention in that they all relate to artificial intelligence, but they differ from the present invention in that they do not use multiple artificial intelligences. Prior art literature

[0007] Prior Art 001: KR 10-2023-0074663 A (Published May 31, 2023) Prior Art 002: KR 10-2023-0063276 A (Published May 9, 2023) Prior Art 003: KR ​​10-2022-0086872 A (Published June 24, 2022) Prior Art 004: KR 10-2021-0084846 A (Published July 8, 2021) The problem to be solved

[0008] The present invention relates to a method for using artificial intelligence multi-model integration. means of solving the problem

[0010] The present invention has been devised to solve the problems of the prior art as described above, and comprises: a model list output step (S100) in which a list of AI models available for use is transmitted from a server (10) to a user terminal (20) and the AI ​​model list received by the user terminal (20) is output; after the model list output step (S100), a selection option transmission step (S200) in which, when a user selects an AI model to use through the user terminal (20) and inputs data, the selected AI model and the input data are transmitted from the user terminal (20) to the server (10); after the selection option transmission step (S200), a preprocessing step (S300) in which the data received by the server (10) is preprocessed; after the preprocessing step (S300), an execution variable setting step (S400) in which the execution variables of the AI ​​model are set by the server (10); after the execution variable setting step (S400), a model execution step (S500) in which the selected AI model is executed by the server (10) based on the data input by the user to obtain a result value; and the model After the execution step (S500), the method includes a result output step (S600) for outputting the result value obtained from the server (10) to the user terminal (20).

[0011] The present invention is devised to solve the problems of the prior art as described above, and includes an accuracy judgment step (S700) for determining the accuracy of a result value in the server (10) after the model execution step (S500) and before the result output step (S600).

[0012] The present invention is devised to solve the problems of the prior art as described above, wherein in the model list output step (S100), the number of AI models included in the list of available AI models transmitted by the server (10) to the user terminal (20) is at least two, and the user selects at least two AI models to use through the user terminal (20), and includes an integration method candidate transmission step (S310) in which the server (10) transmits an AI model integration method candidate to the user terminal (20) after the preprocessing step (S300) and before the execution variable setting step (S400).

[0013] The present invention has been devised to solve the problems of the prior art as described above, and the model list output step (S100) outputs sample data regarding the expected result when using the artificial intelligence model selected by the user, along with the artificial intelligence model list.

[0014] The present invention has been devised to solve the problems of the prior art as described above, and the model list output step (S100) outputs sample data regarding the expected result when a plurality of artificial intelligence models selected by a user are combined and used, along with the artificial intelligence model list. Effects of the invention

[0016] According to the method for integrating multiple artificial intelligence models according to various embodiments of the present invention as described above, since multiple artificial intelligence models are used, there is an effect of obtaining more accurate result values.

[0017] In addition, according to the present invention, if it is determined in the accuracy judgment step that the result value has an accuracy below a certain level, the model variables are reset and the artificial intelligence model is re-executed, thereby having the effect of guaranteeing an accuracy above a certain level.

[0018] In addition, according to the present invention, in the model list output step, sample data regarding the expected result value when using a specific artificial intelligence model or a combination of multiple artificial intelligence models is output together to the user terminal, so the user has the effect of being able to select a more suitable artificial intelligence model. Brief explanation of the drawing

[0020] FIG. 1 is a block diagram of a user terminal and a server performing an artificial intelligence multi-model integration usage method according to the present invention. FIG. 2 is a block diagram of a user terminal, a server, and a database for performing an artificial intelligence multi-model integration method according to the present invention. FIG. 3 is a flowchart from the user's perspective of the artificial intelligence multi-model integration usage method according to the present invention. FIG. 4 is a flowchart of the artificial intelligence multi-model integration usage method according to the present invention from the perspective of a user terminal and a server. FIGS. 5 to 10 are flowcharts according to various embodiments of the method for integrating artificial intelligence multi-models according to the present invention. Specific details for implementing the invention

[0021] Hereinafter, the most preferred embodiment of the present invention is described in detail so that a person skilled in the art to which the present invention pertains can easily practice the present invention.

[0022] The numbers cited in the following embodiments are not limited to the subjects of citation but may be applied to all embodiments. Subjects that exhibit the same purpose and effect as the configuration presented in the embodiments correspond to equivalent substitute subjects. The superordinate concepts presented in the embodiments include subjects of subordinate concepts that are not described.

[0024] [Example 1-1] The present invention relates to a method for integrating and using multiple artificial intelligence models, comprising: a model list output step (S100) in which a list of artificial intelligence models available for use is transmitted from a server (10) to a user terminal (20) and the list of artificial intelligence models received by the user terminal (20) is output; after the model list output step (S100), a selection option transmission step (S200) in which, when a user selects an artificial intelligence model to use through the user terminal (20) and inputs data, the selected artificial intelligence model and the input data are transmitted from the user terminal (20) to the server (10); after the selection option transmission step (S200), a preprocessing step (S300) in which the data received by the server (10) is preprocessed; after the preprocessing step (S300), an execution variable setting step (S400) in which the execution variables of the artificial intelligence model are set by the server (10); and after the execution variable setting step (S400), a model execution step (S500) in which the selected artificial intelligence model is executed by the server (10) based on the data input by the user to obtain a result value. After the above model execution step (S500), the method includes a result output step (S600) for outputting the result value obtained from the server (10) to the user terminal (20).

[0025] [Example 1-2] The present invention relates to a method for using an artificial intelligence multi-model integration, and in Example 1-1, after the model execution step (S500), an update step (S510) for storing the result value in a database (30) is included.

[0026] [Example 1-3] The present invention relates to a method for using an artificial intelligence multi-model integration. In Example 1-1, after the option transmission step (S200), the server (10) includes a sample data determination step (S210) for determining whether the data entered by the user is identical to the previously stored sample data.

[0027] [Examples 1-4] The present invention relates to a method for using an artificial intelligence multi-model integration. In Example 1-3, after the sample data judgment step (S210), if it is determined that the data entered by the user is identical to the previously stored sample data, the server (10) transmits the previously stored sample data to the user terminal (20).

[0028] [Examples 1-5] The present invention relates to a method for integrating multiple artificial intelligence models. In Example 1-3, if it is determined in the sample data judgment step (S210) that the data entered by the user is not identical to the previously stored sample data, the preprocessing step (S300) is performed.

[0029] [Examples 1-6] The present invention relates to a method for using an artificial intelligence multi-model integration, and in Example 1-3, the sample data is stored in a database (30).

[0030] [Example 1-7] The present invention relates to a method for integrating multiple artificial intelligence models. In Example 1-1, the model list output step (S100) transmits the artificial intelligence model list along with feature information for each artificial intelligence model from the server (10) to the user terminal (20), and the user terminal (20) outputs the feature information for each artificial intelligence model received from the server (10).

[0031] The present invention relates to a method for integrating and using multiple artificial intelligence models. In recent years, the artificial intelligence market has experienced explosive growth, and driven by this rapid market expansion, numerous companies are jumping into the development of AI technology. To maintain competitiveness, these companies are developing individual AI models in different ways and are opening up some of the resulting products for use in limited spaces and in limited ways. Considering the characteristics of artificial intelligence, where the combination of various data and AI models can lead to the emergence of diverse services and increased model performance, it is difficult to view such a closed environment as having a positive effect on market growth. Furthermore, while AI models are very useful in themselves, they are difficult to access for many people due to the difficulty of training and using them. The method for integrating and using multiple artificial intelligence models according to the present invention is designed to enable users to use various types of AI models more easily.

[0032] To solve the technical problem described above, the method for using artificial intelligence multi-model integration according to the present invention may include a model list output step (S100), an option transmission step (S200), a preprocessing step (S300), an execution variable setting step (S400), a model execution step (S500), and a result output step (S600).

[0033] The model list output step (S100) is a step in which the server (10) transmits a list of artificial intelligence models available to the user to the user terminal (20), and the user terminal (20) outputs the list of artificial intelligence models received from the server (10). The user terminal (20) may be an electronic device capable of connecting to the server (10). For example, the user terminal (20) may be an electronic device such as a smartphone, PC, or tablet. The user terminal (20) may be connected to the server (10) via wired or wireless communication. Here, the wireless communication method may be at least one of a mobile communication network or wireless internet. In the model list output step (S100), the server (10) may transmit a list containing at least two artificial intelligence models to the user terminal (20). The user may select a specific artificial intelligence model from the list of artificial intelligence models output to the user terminal (20). The artificial intelligence models used in the present invention may be artificial intelligence models available to general users in the market, and may be artificial intelligence models such as ChatGPT, Gemini, Stable Diffusion, etc.

[0034] The option transmission step (S200) is a step in which, when a user selects an artificial intelligence model and inputs data, the user terminal (20) transmits information and data of the artificial intelligence model selected by the user to the server (10). Here, data may be data related to a specific problem that the artificial intelligence model intends to solve. The artificial intelligence model performs tasks such as prediction, classification, generation, and recommendation. In the tasks performed by the artificial intelligence model, the user's requirements may mainly consist of text, but in some cases, there may be requirements such as images, videos, or music. For example, when a user requests the synthesis of two specific images, the data input by the user through the user terminal (20) may be images, and the result may also be images. As another example, when a user requests a recommendation for a specific book, the data input by the user through the user terminal (20) may be text, and the result may also be text.

[0035] The preprocessing step (S300) is a step in which data received from a user terminal (20) is preprocessed at the server (10). Preprocessing is the process of converting data into a form that an artificial intelligence model can understand so that the artificial intelligence model can understand a question and provide an accurate answer. Preprocessing is performed to ensure consistency with the model training data, remove unnecessary information, and convert the input data into a form that the artificial intelligence model can process. For example, if the data entered by the user to the user terminal (20) is in text format, the preprocessing process involves the steps of sentence tokenization, stop word removal, morphological analysis, sentence vectorization, and normalization. Here, sentence tokenization refers to the process of dividing a sentence into words or phrases; stop word removal is the process of removing words that do not add meaning, such as '의', '가', and '는'; morphological analysis is the process of dividing words into the smallest meaningful units; sentence vectorization is the process of converting a sentence into a numerical vector; and normalization is the process of converting various expression methods into a single standard form. This preprocessing is an important step for improving the performance and generalization capabilities of artificial intelligence models, and different preprocessing methods are used depending on the type of AI model.

[0036] The execution variable setting step (S400) sets the execution variables of the artificial intelligence model on the server (10). Execution variables of the artificial intelligence model refer to various setting values ​​that affect the result when the artificial intelligence model performs a specific task. The types of execution variables of the artificial intelligence model may include learning rate, epoch, batch size, activation function, optimizer, regularization, dropout, etc.

[0037] The model execution step (S500) executes an artificial intelligence model based on data input by a user on the server (10) to obtain a result value.

[0038] The result output step (S600) is a step in which the result value obtained through the model execution step (S500) is transmitted from the server (10) to the user terminal (20), and the user terminal (20) outputs the received result value.

[0039] The method for using artificial intelligence multi-model integration according to the present invention may include an update step (S510) performed after a model execution step (S500).

[0040] The update step (S510) stores the result value in the database (30).

[0041] The database (30) may include separate sample data. Sample data refers to data regarding examples of what kind of results are produced when a specific artificial intelligence model is used when a user checks the list of artificial intelligence models displayed on the user terminal (20), and may be displayed together with the list of artificial intelligence models. In some cases, the user may want a result value of the same type as this sample data. Accordingly, the method for using multiple artificial intelligence models in an integrated manner according to the present invention may further include a sample data judgment step (S210) in which, after the option transmission step (S200), the server (10) determines whether the data entered by the user is the same as the previously stored sample data.

[0042] The pre-stored sample data used in the sample data judgment step (S210) can be stored in the database (30). That is, in the sample data judgment step (S210), the server (10) determines whether the data entered by the user matches the sample data stored in the database (30) through mutual communication with the database (30). If it is determined in the sample data judgment step (S210) that the data entered by the user is identical to the sample data, the server (10) transmits the sample data pre-stored in the database (30) from the server (10) to the user terminal (20) through the sample data transmission step (S220), so that the sample data is output from the user terminal (20). However, if it is determined in the sample data judgment step (S210) that the data entered by the user is not identical to the sample data pre-stored in the database (30), a preprocessing step (S300) may be performed.

[0043] In the model list output step (S100), the user must check the list of artificial intelligence models displayed on the user terminal (20) and select a specific artificial intelligence model. The name of each artificial intelligence model is displayed in this list of artificial intelligence models; however, if only the name of the artificial intelligence model is displayed, it is difficult for a user who does not primarily use artificial intelligence models to know which artificial intelligence model they want. For example, a user who wants to generate a specific image must select an artificial intelligence model that generates images, but it is difficult to determine which of the artificial intelligence models displayed on the user terminal (20) is the model that generates images. Therefore, in the model list output step (S100), the server (10) can transmit the artificial intelligence model list along with the characteristics of each artificial intelligence model to the user terminal (20), and the user terminal (20) can output the artificial intelligence model list received from the server (10) along with the characteristic information for each artificial intelligence model, thereby providing additional information to the user in selecting an artificial intelligence model.

[0044] In addition, unlike this method, in the model list output step (S100), the user terminal (20) may provide a category of artificial intelligence models, and when the user selects a specific category, a method of outputting a list of artificial intelligence models corresponding to that type may also be used.

[0046] [Example 2-1] The present invention relates to a method for integrating multiple artificial intelligence models. In Example 1-1, after the model execution step (S500) and before the result output step (S600), the server (10) includes an accuracy judgment step (S700) for determining the accuracy of the result value.

[0047] [Example 2-2] The present invention relates to a method for using an integrated artificial intelligence multi-model. In Example 2-1, if the server (10) of the accuracy judgment step (S700) determines that the accuracy of the result value is below a reference value, it re-executes the model execution variable setting step (S400), the model execution step (S500), and the accuracy judgment step (S700).

[0048] [Example 2-3] The present invention relates to a method for using an integrated artificial intelligence multi-model. In Example 2-1, if the server (10) determines in the accuracy judgment step (S700) that the accuracy of the result value is greater than or equal to a reference value, it performs the result output step (S600).

[0049] The present invention may further include an accuracy determination step (S700). The accuracy determination step (S700) is a step performed after the model execution step (S500) and determines the accuracy of the result value resulting from the execution of the artificial intelligence model at the server (10) before the result output step (S600).

[0050] In the accuracy judgment step (S700), if the server (10) determines that the accuracy of the execution result of the artificial intelligence model is below a threshold, it can re-execute the model execution variable setting step (S400), the model execution step (S500), and the accuracy judgment step (S700) to increase the accuracy of the result. This is because the accuracy may vary depending on the execution variables set in the model execution variable setting step (S400).

[0051] However, in the accuracy judgment step (S700), if the server (10) determines that the accuracy of the execution result value of the artificial intelligence model exceeds a threshold, it may perform the result output step (S600) to output the result value to the user terminal (20).

[0053] [Example 3-1] The present invention relates to a method for using multiple artificial intelligence models in an integrated manner. In Example 1-1, the number of artificial intelligence models included in the list of available artificial intelligence models transmitted by the server (10) to the user terminal (20) in the model list output step (S100) is at least two, and the user selects at least two artificial intelligence models to use through the user terminal (20). The method includes an integration method candidate transmission step (S310) in which the server (10) transmits a candidate for an artificial intelligence model integration method to the user terminal (20) after the preprocessing step (S300) and before the execution variable setting step (S400).

[0054] [Example 3-2] The present invention relates to a method for using artificial intelligence multi-model integration. In Example 3-1, the candidate for the integration method in the transmission step (S310) includes at least one of a pipeline method, a merge method, a voting method, an average method, a weighting method, and a stacking method.

[0055] [Example 3-3] The present invention relates to a method for using artificial intelligence multi-model integration. In Example 3-1, after the integration method candidate transmission step (S310) and before the execution variable setting step (S400), if a user selects any one of the artificial intelligence model integration method candidates, the method includes a model integration step (S320) for integrating at least two artificial intelligence models using the selected artificial intelligence model integration method.

[0056] [Example 3-4] The present invention relates to a method for using artificial intelligence multi-model integration. In Example 3-1, after the preprocessing step (S300) and before the integration method candidate transmission step (S310), the method includes a model integration possibility determination step (S330) for determining whether the artificial intelligence model selected by the user at the server (10) is a combination that can be integrated.

[0057] [Example 3-5] The present invention relates to a method for using an artificial intelligence multi-model integration. In Example 3-4, in the step of determining whether the model integration is possible (S330), if the server (10) determines that the artificial intelligence model selected by the user is not a combination that can be integrated, it transmits a signal to the user terminal (20) indicating that integration is impossible.

[0058] [Example 3-6] The present invention relates to a method for using an artificial intelligence multi-model integration. In Example 3-4, in the step of determining whether the model integration is possible (S330), if the server (10) determines that the artificial intelligence model selected by the user is integrable, it performs the step of transmitting the integration method candidate (S310).

[0059] The user can select multiple artificial intelligence models included in the list of artificial intelligence models output to the user terminal (20) of the model list output step (S100). That is, in the present invention, multiple artificial intelligence models can be used together to obtain a better result value.

[0060] In order to use multiple artificial intelligence models, a method for integrating multiple artificial intelligence models is selected. To this end, the method for integrating multiple artificial intelligence models according to the present invention may include a step of transmitting integration method candidates (S310) performed between a preprocessing step (S300) and an execution variable setting step (S400).

[0061] The integration method candidate transmission step (S310) is a step of transmitting candidate information of an artificial intelligence model integration method from the server (10) to the user terminal (20). The candidate information of the artificial intelligence model integration method may include a pipeline method, a merge method, a voting method, an averaging method, a weighted averaging method, a stacking method, etc. However, the multiple candidates of artificial intelligence model integration methods used in the present invention are not limited to those described above, and other artificial intelligence model integration methods may be used in addition to the methods described above.

[0062] The pipeline approach is a method that performs a single complex task by sequentially connecting multiple artificial intelligence models. In a pipeline, each AI model is responsible for a specific sub-task at a particular stage, and the structure can be configured such that the output of the previous model serves as the input for the next. The pipeline approach is characterized by its suitability for solving complex problems.

[0063] The merge method refers to combining multiple artificial intelligence (AI) models into a single one, a process that creates a new model by combining the parameters or output values ​​of each model. When merging multiple AI models to create a new model, the average of the parameters from each model may be used. The merge method has the effect of reducing the size of the AI ​​model, lowering computational costs, and achieving higher performance by combining the strengths of individual models.

[0064] The aforementioned voting method, averaging method, weighting method, and stacking method are methods for integrating artificial intelligence models to process result values.

[0065] When a user selects one of the candidates for artificial intelligence model integration methods displayed on the user terminal (20), in the model integration step (S320) performed after the integration method candidate transmission step (S310), the selected multiple artificial intelligence models are integrated using the selected method, and then the execution variable setting step (S400) is performed.

[0066] In some cases, multiple AI models selected by the user in the model list output step (S100) may be a combination that is difficult to integrate with each other. For example, if the user selects Stable Diffusion, which primarily generates images, and ChatGPT, which outputs answers to questions in text format, these two AI models may not be integrated because their output methods and execution methods are different. Therefore, the server (10) can transmit a signal to the user terminal (20) indicating that integration is impossible if the multiple AI models selected by the user are a combination that cannot be integrated with each other, through the preprocessing step (S300) and the integration method candidate transmission model integration feasibility determination step (S330). If the server (10) determines in the model integration feasibility determination step (S330) that the AI ​​models selected by the user are a combination that can be integrated, the integration method candidate transmission step (S310) can be performed.

[0068] [Example 4-1] The present invention relates to a method for integrating multiple artificial intelligence models. In Example 1-1, the model list output step (S100) outputs sample data regarding the expected result when using an artificial intelligence model selected by the user, along with the artificial intelligence model list.

[0069] [Example 4-2] The present invention relates to a method for using an artificial intelligence multi-model integration, wherein in Example 4-1, the sample data is stored in a database (30).

[0070] The artificial intelligence models used in the present invention each have specific characteristics. Taking an artificial intelligence model that primarily generates images as an example, DALL-E 2 developed by OpenAI has the characteristic of being able to generate very realistic images and imitate various art styles. In contrast, an artificial intelligence model called Midjourney has the characteristic of being capable of high-level artistic expression. However, it may be difficult for the user to understand the characteristics of these artificial intelligence models. In order to make the user easily understand the characteristics of these artificial intelligence models, in the model list output step (S100), the server (10) transmits sample data regarding the expected result when using the artificial intelligence model selected by the user to the user terminal (20) along with the artificial intelligence model list, and the user terminal (20) can output the artificial intelligence model list received from the server (10) and the sample data regarding the expected result when using a specific artificial intelligence model. Through the sample data output to the user terminal (20), the user can more easily select the artificial intelligence model they want.

[0071] The above-described sample data may be stored in a database (30), and the server (10) may transmit the sample data to the user terminal (20) through communication with the database (30).

[0073] [Example 5-1] The present invention relates to a method for integrating multiple artificial intelligence models. In Example 3-1, the model list output step (S100) outputs sample data regarding the expected result when a plurality of artificial intelligence models selected by a user are combined and used, along with the artificial intelligence model list.

[0074] [Example 5-2] The present invention relates to a method for using an artificial intelligence multi-model integration, wherein in Example 5-1, the sample data is stored in a database (30).

[0075] Sample data regarding the expected results when using the artificial intelligence models described in Examples 4-1 to 4-2 above also applies when the user combines and uses multiple artificial intelligence models. That is, in the model list output step (S100), the server (10) transmits sample data regarding the expected results when the user combines and uses multiple artificial intelligence models selected by the user to the user terminal (20), and the user terminal (20) can output the received sample data together with the artificial intelligence model list.

[0076] However, since the number of combinations of these selected multiple artificial intelligence models increases as the number of artificial intelligence models increases, there may be a problem with insufficient transmission speed and storage space of the user terminal (20) to transmit all combination sample data from the server (10) to the user terminal (20) at once. Therefore, when a user preliminarily selects multiple artificial intelligence models on the user terminal (20) in the model list output step (S100), the user terminal (20) transmits the information of the multiple artificial intelligence models selected by the user to the server (10), and the server (10) can receive sample data of the combination of multiple artificial intelligence models selected by the user from the database (30) and transmit it to the user terminal (20), and the user terminal (20) can output the sample data received from the server (10).

[0078] The present invention is not limited to the embodiments described above and has a diverse scope of application. Furthermore, it is understood that anyone with ordinary knowledge in the field to which the present invention pertains can make various modifications without departing from the essence of the invention as claimed in the claims. Explanation of the symbols

[0080] 10: Server 20: User terminal 30: Database S100: Model list output step S200: Optional transmission step S210: Sample data judgment step S220: Sample data transmission step S300: Preprocessing step S310: Integration method candidate transmission stage S320: Model Integration Phase S330: Model integration feasibility determination stage S400: Execution variable setting step S500: Model execution phase S510: Update phase S600: Result output stage S700: Accuracy judgment stage

Claims

Claim 1 A model list output step (S100) in which a list of AI models available for use is transmitted from a server (10) to a user terminal (20) and the AI ​​model list received by the user terminal (20) is output; after the model list output step (S100), when a user selects an AI model to use through the user terminal (20) and inputs data, the user terminal (20) transmits the selected AI model and the input data to the server (10); after the option transmission step (S200), a preprocessing step (S300) in which the data received by the server (10) is preprocessed; after the preprocessing step (S300), an execution variable setting step (S400) in which the execution variables of the AI ​​model are set by the server (10); after the execution variable setting step (S400), a model execution step (S500) in which the selected AI model is executed by the server (10) based on the data input by the user to obtain a result value; after the model execution step (S500), the result obtained by the server (10) The method includes a result output step (S600) for outputting a result value to the user terminal (20); wherein, in the model list output step (S100), the number of AI models included in the list of available AI models transmitted by the server (10) to the user terminal (20) is at least two, and the user selects at least two AI models to use through the user terminal (20); and, after the preprocessing step (S300) and before the execution variable setting step (S400), the method includes an integration method candidate transmission step (S310) for transmitting an AI model integration method candidate from the server (10) to the user terminal (20); and, after the integration method candidate transmission step (S310) and before the execution variable setting step (S400), if the user selects any one of the AI ​​model integration method candidates, the method includes a model integration step (S320) for integrating at least two AI models using the selected AI model integration method.A method for using artificial intelligence multi-model integration including