KR102596930B1 - Method, device and system for automating creation of homepage production information using artificial intelligence model - Google Patents

Abstract

According to one embodiment, a method of automating the creation of homepage production information using an artificial intelligence model, performed by a device, includes receiving industry information indicating the industry of a first company from a first user terminal; Generating a first input signal by encoding the industry information; The first input signal is input to a first artificial intelligence model learned to recommend the design of the homepage according to the industry, and when the recommended design is selected through the first input signal, the recommended design is selected from the first artificial intelligence model. obtaining a first output signal representing; The first input signal is input to a second artificial intelligence model trained to recommend menus of the homepage according to the industry, and when a recommended menu is selected through the first input signal, the recommended menu is selected from the second artificial intelligence model. obtaining a second output signal representing; creating a first homepage based on the recommended design and the recommended menu; A method of automating the creation of homepage production information using an artificial intelligence model is provided, including the step of transmitting the production information of the first homepage to the first user terminal.

Description

Method, device, and system for automating homepage production information creation using artificial intelligence models {METHOD, DEVICE AND SYSTEM FOR AUTOMATING CREATION OF HOMEPAGE PRODUCTION INFORMATION USING ARTIFICIAL INTELLIGENCE MODEL}

The examples below relate to technology for automating the creation of homepage production information using an artificial intelligence model.

For the average user, creating a homepage is not an easy task. Conventionally, homepages were created using the HTML (Hypertext Markup Language) language, but since it is difficult for general users to learn the HTML language, tools are being proposed to create homepages more conveniently.

However, even if a homepage creation tool is used, it is difficult for users to create a homepage themselves without the help of a developer.

Therefore, there is a need for research and development on technology that can automatically create homepages and increase user convenience.

Korean Patent No. 10-2420825 Korean Patent Publication No. 10-2021-0156366 Korean Patent No. 10-1619845 Korean Patent Publication No. 10-2015-0102150

According to one embodiment, the purpose is to provide a method, device, and system that automates the creation of homepage production information using an artificial intelligence model.

The object of the present invention is not limited to the object mentioned above, and other objects not mentioned can be clearly understood from the description below.

According to one embodiment, a method of automating the creation of homepage production information using an artificial intelligence model, performed by a device, includes receiving industry information indicating the industry of a first company from a first user terminal; Generating a first input signal by encoding the industry information; The first input signal is input to a first artificial intelligence model learned to recommend the design of the homepage according to the industry, and when the recommended design is selected through the first input signal, the recommended design is selected from the first artificial intelligence model. obtaining a first output signal representing; The first input signal is input to a second artificial intelligence model trained to recommend menus of the homepage according to the industry, and when a recommended menu is selected through the first input signal, the recommended menu is selected from the second artificial intelligence model. obtaining a second output signal representing; creating a first homepage based on the recommended design and the recommended menu; A method of automating the creation of homepage production information using an artificial intelligence model is provided, including the step of transmitting the production information of the first homepage to the first user terminal.

The method of automating the creation of homepage production information using the artificial intelligence model includes, after transmitting the production information of the first homepage to the first user terminal, the design and menu of the first homepage from the first user terminal. Upon receiving a confirmation request, confirming the recommended design as a design of the first homepage and confirming the recommended menu as a menu of the first homepage; If it is confirmed that information about the first item is needed to create the contents of the first menu among the menus of the first homepage, based on the pre-registered first company information, extracting first item information, which is information; Generating matching information by matching the first menu and the first item information; Generating a second input signal by encoding the matching information; The second input signal is input to a third artificial intelligence model learned to answer the contents of the menu according to the menu and item, and when a recommended answer is generated through the second input signal, the recommendation is made from the third artificial intelligence model. obtaining a third output signal representing an answer; Creating a first page on which contents of the first menu are displayed based on the recommended answer; And it may further include transmitting production information of the first page to the first user terminal.

The method of automating the creation of homepage production information using the artificial intelligence model includes, after transmitting the production information of the first homepage to the first user terminal, the design and menu of the first homepage from the first user terminal. Upon receiving a modification request, modifying at least one of the recommended design and the recommended menu according to the modification request for the design and menu of the first homepage; When the recommended design is modified, a first learning signal is generated indicating that the recommended design selected through the first input signal has been modified to another design, and the first learning signal is input to the first artificial intelligence model. , processing the first artificial intelligence model to be learned and updated through the first learning signal; And when the recommended menu is modified, generate a second learning signal indicating that the recommended menu selected through the first input signal has been modified to another menu, and input the second learning signal to the second artificial intelligence model. Thus, the step of processing the second artificial intelligence model to be learned and updated through the second learning signal may be further included.

The method of automating the creation of homepage production information using the artificial intelligence model includes, after transmitting the production information of the first page to the first user terminal, a request for modification of the contents of the first page from the first user terminal. Upon receiving, modifying the content of the first page according to a request for modification of the content of the first page; And when the content of the first page is modified, generate a third learning signal indicating that the recommended answer generated through the second input signal has been modified to different content, and apply the third learning signal to the third artificial intelligence model. It may further include inputting a signal and processing the third artificial intelligence model to be learned and updated through the third learning signal.

According to one embodiment, by selecting recommended designs and recommended menus using an artificial intelligence model and creating a homepage based on the recommended designs and recommended menus, user convenience can be increased by automating the creation of the homepage. .

Meanwhile, the effects according to the embodiments are not limited to those mentioned above, and other effects not mentioned can be clearly understood by those skilled in the art from the description below.

1 is a diagram schematically showing the configuration of a system according to an embodiment.
Figure 2 is a diagram showing the configuration of the operation of the system according to one embodiment.
Figure 3 is a flow chart to explain the process of automating the creation of homepage production information using an artificial intelligence model.
Figure 4 is a flowchart for explaining the process of creating a web page on which menu contents are displayed according to one embodiment.
Figure 5 is a flow chart to explain the process of learning modifications to the design and menu according to an embodiment.
Figure 6 is a flowchart for explaining a process of learning modifications to menu contents according to an embodiment.
Figure 7 is a flow chart to explain the process of selecting a recommended design according to an embodiment.
Figure 8 is a flowchart for explaining the process of selecting a recommended menu according to an embodiment.
9 to 11 are flowcharts for explaining the process of setting the background color of the homepage according to one embodiment.
Figure 12 is an exemplary diagram of the configuration of a device according to an embodiment.

Hereinafter, embodiments will be described in detail with reference to the attached drawings. However, various changes can be made to the embodiments, so the scope of the patent application is not limited or limited by these embodiments. It should be understood that all changes, equivalents, or substitutes for the embodiments are included in the scope of rights.

Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only and may be modified and implemented in various forms. Accordingly, the embodiments are not limited to the specific disclosed form, and the scope of the present specification includes changes, equivalents, or substitutes included in the technical spirit.

Terms such as first or second may be used to describe various components, but these terms should be interpreted only for the purpose of distinguishing one component from another component. For example, a first component may be named a second component, and similarly, the second component may also be named a first component.

When a component is referred to as being “connected” to another component, it should be understood that it may be directly connected or connected to the other component, but that other components may exist in between.

The terms used in the examples are for descriptive purposes only and should not be construed as limiting. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as “comprise” or “have” are intended to designate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, but are not intended to indicate the presence of one or more other features. It should be understood that this does not exclude in advance the possibility of the existence or addition of elements, numbers, steps, operations, components, parts, or combinations thereof.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by a person of ordinary skill in the technical field to which the embodiments belong. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and unless explicitly defined in the present application, should not be interpreted in an ideal or excessively formal sense. No.

In addition, when describing with reference to the accompanying drawings, identical components will be assigned the same reference numerals regardless of the reference numerals, and overlapping descriptions thereof will be omitted. In describing the embodiments, if it is determined that detailed descriptions of related known technologies may unnecessarily obscure the gist of the embodiments, the detailed descriptions are omitted.

Embodiments may be implemented in various types of products such as personal computers, laptop computers, tablet computers, smart phones, televisions, smart home appliances, intelligent vehicles, kiosks, and wearable devices.

In an embodiment, an artificial intelligence (AI) system is a computer system that implements human-level intelligence, and unlike existing rule-based smart systems, it is a system in which machines learn and make decisions on their own. As artificial intelligence systems are used, recognition rates improve and sellers' tastes can be more accurately understood, and existing rule-based smart systems are gradually being replaced by deep learning-based artificial intelligence systems.

Artificial intelligence technology consists of machine learning and element technologies using machine learning. Machine learning is an algorithmic technology that classifies/learns the characteristics of input data on its own, and elemental technology is a technology that mimics the functions of the human brain such as cognition and judgment by utilizing machine learning algorithms such as deep learning, including linguistic understanding and visual It consists of technical areas such as understanding, reasoning/prediction, knowledge expression, and motion control.

The various fields where artificial intelligence technology is applied are as follows. Linguistic understanding is a technology that recognizes and applies/processes human language/characters and includes natural language processing, machine translation, conversation systems, question and answer, and voice recognition/synthesis. Visual understanding is a technology that recognizes and processes objects like human vision, and includes object recognition, object tracking, image search, person recognition, scene understanding, spatial understanding, and image improvement. Inferential prediction is a technology that judges information to make logical inferences and predictions, and includes knowledge/probability-based reasoning, optimization prediction, preference-based planning, and recommendations. Knowledge expression is a technology that automatically processes human experience information into knowledge data, and includes knowledge construction (data creation/classification) and knowledge management (data utilization). Motion control is a technology that controls the autonomous driving of vehicles and the movement of robots, and includes motion control (navigation, collision, driving), operation control (behavior control), etc.

Generally, in order to apply machine learning algorithms to real life, learning is performed using a trial and error method due to the nature of the basic methodology of machine learning. In particular, deep learning requires hundreds of thousands of iterations. It is impossible to execute this in an actual physical external environment, so instead, the actual physical external environment is virtually implemented on a computer and learning is performed through simulation.

1 is a diagram schematically showing the configuration of a system according to an embodiment.

Referring to FIG. 1, a system according to an embodiment may include a plurality of user terminals 100 and devices 200 that can communicate with each other through a communication network.

First, a communication network can be configured regardless of the communication mode, such as wired or wireless, and can be implemented in various forms to enable communication between servers and between servers and terminals.

Each of the plurality of user terminals 100 may be implemented as a computing device with a communication function, and may be implemented as, for example, a mobile phone, a desktop PC, a laptop PC, a tablet PC, a smartphone, etc., but is not limited thereto. It can also be implemented as various types of communication devices that can be connected to external servers.

The plurality of user terminals 100 are terminals used by representatives of each company. The first user terminal 110 is used by the first user, who is in charge of the first company, and the second user is used by the representative of the second company. It may include a second user terminal 120, etc.

Each of the plurality of user terminals 100 may be configured to perform all or part of the calculation function, storage/reference function, input/output function, and control function of a typical computer. A plurality of user terminals 100 may be configured to communicate with the device 200 in a wired or wireless manner.

Each of the plurality of user terminals 100 is connected to a web page operated by a person or organization providing services using the device 200, or is developed and distributed by a person or organization providing services using the device 200. An application can be installed. Each of the plurality of user terminals 100 may be linked to the device 200 through a web page or application.

Each of the plurality of user terminals 100 may access the device 200 through a web page or application provided by the device 200.

The device 200 may be its own server owned by a person or organization that provides services using the device 200, a cloud server, or a p2p (peer-to-peer) set of distributed nodes. It may be possible. The device 200 may be configured to perform all or part of the calculation function, storage/reference function, input/output function, and control function of a typical computer. The device 200 may be equipped with at least one artificial intelligence model that performs an inference function.

The device 200 may be configured to communicate wired or wirelessly with a plurality of user terminals 100, control the operation of each of the plurality of user terminals 100, and display certain information on the screen of each of the plurality of user terminals 100. You can control whether to display it or not.

The device 200 is implemented as a server that provides a homepage production automation service and can provide a platform for homepage production.

Meanwhile, for convenience of explanation, only the first user terminal 110 and the second user terminal 120 are shown in FIG. 1 among the plurality of user terminals 100, but the number of terminals may vary depending on the embodiment. . As long as the processing capacity of the device 200 allows, the number of terminals is not particularly limited.

According to one embodiment, the device 200 may select a homepage design and menu based on artificial intelligence and provide a homepage created based on the selected design and menu to the user.

In the present invention, artificial intelligence (AI) refers to technology that imitates human learning ability, reasoning ability, perception ability, etc. and implements this with a computer, and includes concepts such as machine learning and symbolic logic. may include. Machine Learning (ML) is an algorithmic technology that classifies or learns the characteristics of input data on its own. Artificial intelligence technology is a machine learning algorithm that analyzes input data, learns the results of the analysis, and makes judgments or predictions based on the results of the learning. Additionally, technologies that mimic the functions of the human brain, such as cognition and judgment, using machine learning algorithms can also be understood as the category of artificial intelligence. For example, technical fields such as verbal understanding, visual understanding, reasoning/prediction, knowledge representation, and motion control may be included.

Machine learning can refer to the process of training a neural network model using experience processing data. Machine learning can mean that computer software improves its own data processing capabilities. A neural network model is built by modeling the correlation between data, and the correlation can be expressed by a plurality of parameters. A neural network model extracts and analyzes features from given data to derive correlations between data. Repeating this process to optimize the parameters of the neural network model can be called machine learning. For example, a neural network model can learn the mapping (correlation) between input and output for data given as input-output pairs. Alternatively, a neural network model may learn the relationships by deriving regularities between given data even when only input data is given.

An artificial intelligence learning model or neural network model may be designed to implement the human brain structure on a computer, and may include a plurality of network nodes with weights that simulate neurons of a human neural network. A plurality of network nodes may have a connection relationship with each other by simulating the synaptic activity of neurons in which neurons exchange signals through synapses. In an artificial intelligence learning model, multiple network nodes are located in layers of different depths and can exchange data according to convolutional connection relationships. The artificial intelligence learning model may be, for example, an artificial neural network model (Artificial Neural Network), a convolution neural network (CNN) model, etc. As an example, an artificial intelligence learning model may be machine-learned according to methods such as supervised learning, unsupervised learning, and reinforcement learning. Machine learning algorithms for performing machine learning include Decision Tree, Bayesian Network, Support Vector Machine, Artificial Neural Network, and Ada-boost. , Perceptron, Genetic Programming, Clustering, etc. can be used.

Among them, CNN is a type of multilayer perceptrons designed to use minimal preprocessing. CNN consists of one or several convolution layers and general artificial neural network layers on top of them, and additionally utilizes weights and pooling layers. Thanks to this structure, CNN can fully utilize input data with a two-dimensional structure. Compared to other deep learning structures, CNN shows good performance in both video and audio fields. CNNs can also be trained via standard back propagation. CNNs have the advantage of being easier to train and using fewer parameters than other feedforward artificial neural network techniques.

Convolutional networks are neural networks that contain sets of nodes with bound parameters. The increasing size of available training data and the availability of computational power, combined with algorithmic advances such as piecewise linear unit and dropout training, have led to significant improvements in many computer vision tasks. For extremely large data sets, such as those available for many tasks today, overfitting is not critical, and increasing the size of the network improves test accuracy. Optimal use of computing resources becomes a limiting factor. For this purpose, distributed, scalable implementations of deep neural networks can be used.

Figure 2 is a diagram showing the configuration of the operation of the system according to one embodiment.

In FIG. 2, a user can perform an operation through a user terminal, and an operation performed on a server can be performed on the device 200.

Referring to Figure 2, first, the user can input the industry into the user terminal, the server can select a design and menu according to the industry through artificial intelligence, and create a homepage based on the selected design and menu. A detailed explanation related to homepage production will be described later with reference to FIG. 3.

Afterwards, the user can confirm the design and menu of the homepage, and the server classifies the information confirmed by the user by industry on the server, uploads and learns the classified information, creates menu content through artificial intelligence, and creates menu content. Based on this, you can create a web page that displays menu contents. Here, the web page is a page connected to the selection of a menu within the homepage, and a detailed explanation related to web page production will be described later with reference to FIG. 4.

According to one embodiment, the server utilizes learned artificial intelligence to automatically recommend designs and menus using a database based on the industry selected or entered by a new user, and creates a homepage based on the recommended designs and menus. You can. Here, the database stores information on the design, menu structure, and menu contents of homepages created for each industry.

If the user modifies the recommended design and menu, the artificial intelligence can learn by inputting the correction information indicating the part that the user modified into the artificial intelligence that selects the design and menu. A detailed explanation regarding the learning of artificial intelligence for selecting designs and menus will be described later with reference to FIG. 5.

The server can utilize learned artificial intelligence to automatically generate menu contents determined by the user, such as greetings and business information.

If the menu content is modified by the user, the artificial intelligence can be trained by inputting correction information indicating the part modified by the user into the artificial intelligence that generates the menu content. A detailed explanation regarding the learning of artificial intelligence that generates menu contents will be described later with reference to FIG. 6.

Figure 3 is a flow chart to explain the process of automating the creation of homepage production information using an artificial intelligence model.

Referring to FIG. 3, first, in step S301, the device 200 may receive industry information indicating the industry of the first company from the first user terminal 110.

Specifically, the first user, who is in charge of the first company, may input industry information indicating the industry of the first company into the first user terminal 110 in order to request a homepage production request, and the industry information may be entered into the first user terminal 110. When input is input to the terminal 110, the device 200 can receive a homepage creation request request and industry information from the first user terminal 110.

In step S302, the device 200 may generate a first input signal by encoding industry information.

Specifically, the device 200 may perform normal information processing on industry information to generate a first input signal in order to use industry information as input to an artificial intelligence model.

In step S303, the device 200 may input a first input signal to the first artificial intelligence model learned in advance. Here, the first artificial intelligence model is trained to recommend the design of the homepage according to the industry, and may be learned through information about the design of the homepage produced for each industry.

The first artificial intelligence model may be an algorithm that receives an input signal generated by encoding industry information, selects a recommended design through the input signal, and outputs an output signal representing the selected recommended design.

That is, the first artificial intelligence model can consider the company's industry, select a recommended design according to the industry, and output an output signal representing the selected recommended design. A detailed description of the process of selecting a recommended design in the first artificial intelligence model will be described later with reference to FIG. 7.

In step S304, when a recommended design is selected through a first input signal, the device 200 may obtain a first output signal representing the recommended design from the first artificial intelligence model. At this time, the recommended design may be composed of a template in which at least one image to be used on the homepage is arranged, and the first output signal may include information representing the recommended design and information necessary to set the recommended design.

For example, the device 200 inputs a first input signal to the first artificial intelligence model, and when the recommended design is set to the first design through the first input signal, the device 200 represents the first design from the first artificial intelligence model. A first output signal can be obtained, and when the recommended design is set to the second design through the first input signal, a first output signal representing the second design can be obtained from the first artificial intelligence model.

That is, the first artificial intelligence model can select a recommended design through an input signal and output an output signal representing the selected recommended design. To this end, the first artificial intelligence model may be trained in advance to analyze which design to recommend according to the industry, based on information about the design of the homepage created for each industry stored in the database.

The learning device in which the first artificial intelligence model is learned may be the same device as the device 200 that selects the recommended design of the homepage using the learned first artificial intelligence model, or may be a separate device. Below, the process by which the first artificial intelligence model is learned is explained.

First, the learning device can generate input based on industry information.

Specifically, the learning device may perform a preprocessing process on industry information. The pre-processed industry information can be used as input to the first artificial intelligence model, or the input can be generated through normal processing to remove unnecessary information.

Next, the learning device can apply input to the first artificial intelligence model. The first artificial intelligence model may be an artificial neural network learned according to reinforcement learning. The first artificial intelligence model may be a Q-Network, DQN (Depp Q-Network), or relational network (RL) structure suitable for outputting abstract inference through reinforcement learning.

The first artificial intelligence model learned according to reinforcement learning can be updated and optimized by reflecting the evaluation of various rewards.

For example, the first reward can increase the reward value by selecting the design of a homepage classified in the same industry as the recommended design among the pre-made homepages, and the second reward can have the highest average number of visits among the pre-made homepages. The reward value can be increased if the design of many homepages is selected as the recommended design, and the third reward can be increased if the design of the homepage with the longest average access time among pre-made homepages is selected as the recommended design. For the fourth reward, the reward value can be increased if the design of the homepage with the most recent homepage update among pre-made homepages is selected as the recommended design. At this time, information about the pre-made homepages may be stored in the database of the device 200, and the first artificial intelligence model selects among the pre-made homepages based on the information about the pre-made homepages stored in the database. You can select a recommended design.

Next, the learning device may obtain an output from the first artificial intelligence model. At this time, the output of the first artificial intelligence model may be information indicating a recommended design. In other words, the first artificial intelligence model can analyze which design to recommend depending on the industry and output information about the analyzed recommended design.

Next, the learning device may evaluate the output of the first artificial intelligence model and pay a reward.

For example, when the learning device selects the design of a homepage classified into the same industry as the recommended design among pre-made homepages, it awards the first reward and designs the homepage with the highest average number of visits among the pre-made homepages. If a design is selected as the recommended design, a large number of secondary rewards will be awarded. If the design of the homepage with the longest average access maintenance time among pre-made homepages is selected as a recommended design, a large number of 3rd rewards will be awarded, and among pre-made homepages, a large number of secondary rewards will be awarded. If the homepage design with the most recent homepage update is selected as the recommended design, a large number of fourth rewards can be awarded.

Next, the learning device may update the first artificial intelligence model based on the evaluation.

Specifically, the learning device takes actions to be taken in specific states so that the expected value of the sum of rewards is maximized in the environment where the first artificial intelligence model analyzes the recommended design. ) can be updated through the process of optimizing the policy that determines the first artificial intelligence model.

For example, if there is no problem with the result of analyzing the recommended design as the first design according to the first company's industry, the learning device will respond to the result of analyzing the recommended design as the first design according to the first company's industry. When generating first learning data indicating that there is no problem, applying the first learning data to the first artificial intelligence model, and analyzing the recommended design of a company with an industry similar to that of the first company, the recommended design is selected. The first artificial intelligence model can be updated through the process of training the first artificial intelligence model to select a design similar to design 1.

Meanwhile, the process of optimizing the policy can be accomplished through estimating the maximum expected value of the sum of rewards or the maximum value of the Q-function, or estimating the minimum value of the loss function of the Q-function. Estimation of the minimum value of the loss function can be done through stochastic gradient descent (SGD), etc. The process of optimizing the policy is not limited to this, and various optimization algorithms used in reinforcement learning can be used.

The learning device can gradually update the first artificial intelligence model by repeating the learning process of the first artificial intelligence model described above. Through this, the learning device can learn the first artificial intelligence model that analyzes and outputs which design to recommend depending on the industry.

That is, when the learning device analyzes the recommended design of the homepage according to the industry, it can learn the first artificial intelligence model by reflecting reinforcement learning through the first to fourth rewards, etc. and adjusting the analysis standard.

Meanwhile, in step S305, the device 200 may input the first input signal to the second artificial intelligence model learned in advance. Here, the second artificial intelligence model is trained to recommend homepage menus depending on the industry, and may be learned through information about the menus of homepages created for each industry.

The second artificial intelligence model may be an algorithm that receives an input signal generated by encoding industry information, selects a recommended menu through the input signal, and outputs an output signal representing the selected recommended menu.

In other words, the second artificial intelligence model can consider the company's industry, select a recommended menu according to the industry, and output an output signal indicating the selected recommended menu. A detailed description of the process of selecting a recommended menu in the second artificial intelligence model will be described later with reference to FIG. 8.

In step S306, when a recommended menu is selected through the first input signal, the device 200 may obtain a second output signal representing the recommended menu from the second artificial intelligence model. At this time, the recommended menu may be composed of a template in which at least one menu to be used on the homepage is arranged, and the second output signal may include information indicating the recommended menu and information necessary to set the recommended menu.

For example, the device 200 inputs a first input signal to the second artificial intelligence model, and when the recommended menu is set to the first menu through the first input signal, the device 200 displays the first menu from the second artificial intelligence model. A second output signal can be obtained, and when the recommended menu is set to the second menu through the first input signal, a second output signal representing the second menu can be obtained from the second artificial intelligence model.

That is, the second artificial intelligence model can select a recommended menu through an input signal and output an output signal representing the selected recommended menu. To this end, the second artificial intelligence model may be trained in advance to analyze which menu to recommend according to the industry, based on information about the menu of the homepage created for each industry stored in the database.

The learning device in which learning of the second artificial intelligence model is performed may be the same device as the device 200 that selects the recommended menu of the homepage using the learned second artificial intelligence model, or may be a separate device. Below, the process by which the second artificial intelligence model is learned is explained.

First, the learning device can generate input based on industry information.

Specifically, the learning device may perform a preprocessing process on industry information. The pre-processed industry information can be used as input to the second artificial intelligence model, or the input can be generated through normal processing to remove unnecessary information.

Next, the learning device can apply input to the second artificial intelligence model. The second artificial intelligence model may be an artificial neural network learned according to reinforcement learning. The second artificial intelligence model may be a Q-Network, DQN (Depp Q-Network), or relational network (RL) structure suitable for outputting abstract inference through reinforcement learning.

The second artificial intelligence model learned according to reinforcement learning can be updated and optimized by reflecting the evaluation of various rewards.

For example, the 5th reward can increase the reward value by selecting a menu from a homepage classified into the same industry as a recommended menu among the pre-made homepages, and the 6th reward has the highest average number of visits among the pre-made homepages. If menus from many homepages are selected as recommended menus, the reward value can be increased, and for the 7th reward, the reward value can be increased if the menu of the homepage with the longest average access maintenance time among pre-made homepages is selected as recommended menu, For the 8th reward, the reward value can be increased if the menu of the homepage with the most recent homepage update among pre-made homepages is selected as the recommended menu. At this time, information about the pre-made homepages may be stored in the database of the device 200, and the second artificial intelligence model selects among the pre-made homepages based on the information about the pre-made homepages stored in the database. You can select a recommended menu.

Next, the learning device can obtain an output from the second artificial intelligence model. At this time, the output of the second artificial intelligence model may be information indicating a recommended menu. In other words, the second artificial intelligence model can analyze which menu to recommend depending on the industry and output information about the analyzed recommended menu.

Next, the learning device can evaluate the output of the second artificial intelligence model and pay a reward.

For example, if the learning device selects a menu from a homepage classified into the same industry as a recommended menu among pre-made homepages, it awards a large number of fifth rewards, and the menu of the homepage with the highest average number of visits among the pre-made homepages If you select a menu as a recommended menu, many 6th rewards will be awarded. If you select the menu of the homepage with the longest average access time among pre-made homepages as a recommended menu, a lot of 7th rewards will be awarded. Among pre-made homepages, you will receive a lot of 6th rewards. If you select the menu of the homepage with the most recent update as the recommended menu, you can receive the 8th reward in large numbers.

Next, the learning device may update the second artificial intelligence model based on the evaluation.

Specifically, the learning device takes actions to be taken in specific states so that the expected value of the sum of rewards is maximized in the environment where the second artificial intelligence model analyzes the recommended menu. ) can be updated through the process of optimizing the policy that determines the second artificial intelligence model.

For example, if there is no problem with the result of analyzing the recommended menu as the first menu according to the first company's industry, the learning device will respond to the result of analyzing the recommended menu as the first menu according to the first company's industry. When generating second learning data indicating that there is no problem and applying the second learning data to the second artificial intelligence model to analyze the recommended menu of a company with an industry similar to that of the first company, the recommended menu is provided. The second artificial intelligence model can be updated through the process of training the second artificial intelligence model to select a menu similar to menu 1.

Meanwhile, the process of optimizing the policy can be accomplished through estimating the maximum expected value of the sum of rewards or the maximum value of the Q-function, or estimating the minimum value of the loss function of the Q-function. Estimation of the minimum value of the loss function can be done through stochastic gradient descent (SGD), etc. The process of optimizing the policy is not limited to this, and various optimization algorithms used in reinforcement learning can be used.

The learning device can gradually update the second artificial intelligence model by repeating the learning process of the second artificial intelligence model described above. Through this, the learning device can learn a second artificial intelligence model that analyzes and outputs which menu to recommend depending on the industry.

That is, when the learning device analyzes the recommended menu of the homepage according to the industry, it can learn the second artificial intelligence model by reflecting reinforcement learning through the fifth to eighth rewards, etc. and adjusting the analysis standard.

In step S307, the device 200 may create a first homepage based on the recommended design and recommended menu.

Specifically, the device 200 confirms the recommended design through the first output signal, confirms the recommended menu through the second output signal, sets the design of the homepage based on the recommended design, and sets the homepage design based on the recommended menu. You can create the first homepage by setting the menu.

In step S308, the device 200 may transmit production information of the first homepage to the first user terminal 110. Here, the production information of the first homepage may include information necessary to create the first homepage (information on recommended designs, recommended menus, etc.) and results generated by creating the first homepage.

Figure 4 is a flowchart for explaining the process of creating a web page on which menu contents are displayed according to one embodiment.

According to one embodiment, each step shown in FIG. 4 may be performed after step S308.

Referring to FIG. 4, first, in step S401, the device 200 may receive a confirmation request for the design and menu of the first homepage from the first user terminal 110. Here, the confirmation request is a request to confirm the design and menu of the first homepage because there is no part to be modified.

In step S402, the device 200 may confirm the recommended design as the design of the first homepage and the recommended menu as the menu of the first homepage.

In step S403, the device 200 may confirm that information about the first item is needed to create the contents of the first menu among the menus of the first homepage.

For example, if the menu of the first homepage includes menus such as greetings and business information, the device 200 determines that information about the representative item is needed to create the contents of the greeting menu among the menus of the first homepage. You can confirm that information on business performance items is needed to create the contents of the business information menu among the menus on the first homepage.

In step S404, the device 200 may extract first item information, which is information about the first item of the first company, based on pre-registered first company information. For this purpose, the database of the device 200 stores information on a plurality of companies classified by company.

Specifically, the device 200 may query the database to obtain first company information and extract first item information, which is information about the first item of the first company, from the first company information. Here, the first company information may include information about the first company's industry, size, sales, business experience, number of employees, representative, work performance, etc.

For example, if it is confirmed that information about the representative item is needed to create the contents of the greeting menu among the menus of the first homepage, the device 200 provides information about the representative item of the first company based on the first company information. Information can be extracted as first item information.

In step S405, the device 200 may generate matching information by matching the first menu and first item information.

In step S406, the device 200 may generate a second input signal by encoding matching information.

Specifically, in order to use the matching information as input to an artificial intelligence model, the device 200 may perform normal information processing on the matching information and generate a second input signal.

In step S407, the device 200 may input a second input signal to a pre-trained third artificial intelligence model. Here, the third artificial intelligence model is trained to answer the contents of the menu according to the menu and item information, and may be learned through information about the contents of each menu on the homepage produced for each industry.

The third artificial intelligence model may be an algorithm that receives an input signal generated by encoding matching information that matches menu and item information, generates a recommended answer through the input signal, and outputs an output signal representing the generated recommended answer. there is.

That is, the third artificial intelligence model can consider menu and item information, generate a recommended answer to be displayed as the contents of the menu, and output an output signal representing the generated recommended answer.

In step S408, the device 200 may obtain a third output signal indicating a recommended answer from the third artificial intelligence model. At this time, the recommended answer is content to be displayed in the menu, and the third output signal may include information representing the recommended answer.

For example, the device 200 inputs a second input signal to the third artificial intelligence model, and when a recommended answer is generated as the first answer through the second input signal, the device 200 represents the first answer from the third artificial intelligence model. A third output signal can be obtained, and when the recommended answer is generated as a second answer through the second input signal, a third output signal representing the second answer can be obtained from the third artificial intelligence model,

In other words, the third artificial intelligence model can generate a recommended answer to be displayed in the menu through an input signal and output an output signal representing the generated recommended answer. For this purpose, the third artificial intelligence model is pre-trained to analyze which content to generate as an answer according to each menu item on the homepage, based on information about the content of each menu on the homepage created for each industry stored in the database. It may be a state.

The third artificial intelligence model can be implemented as a language model such as BERT, GPT, RoBERTa, etc. to generate recommended answers, and when menu and item information is input as a query, the content to be displayed in the menu can be output as an answer. .

For example, when the greeting menu and representative information are entered as a query, the third artificial intelligence model checks the representative's gender and age based on the representative information and sends the representative's greeting by considering the representative's gender and age. You can create a message and print the generated greeting from the representative as a response.

In step S409, the device 200 may create a first page on which the contents of the first menu are displayed based on the recommended answer.

Specifically, the device 200 may create a first web page connected to the selection of the first menu within the homepage and then set the recommended answer as the content of the first menu to create the first page.

In step S410, the device 200 may transmit production information of the first page to the first user terminal 110. Here, the production information of the first page may include information necessary to produce the first page (information about recommended answers, etc.) and results generated by producing the first page.

Figure 5 is a flow chart to explain the process of learning modifications to the design and menu according to an embodiment.

According to one embodiment, each step shown in FIG. 5 may be performed after step S308.

Referring to FIG. 5, first, in step S501, the device 200 may receive a request to modify the design and menu of the first homepage from the first user terminal 110. Here, the request for modification is a request for modification of at least one of the design and menu of the first homepage because there is a part to be modified.

In step S502, the device 200 may modify at least one of the recommended design and recommended menu according to a request to modify the design and menu of the first homepage.

If the recommended design is modified, steps S503 to S504 may be performed, if the recommended menu is modified, steps S505 to S506 may be performed, and if both the recommended design and the recommended menu are modified, steps S503 Steps S504 to S505 to S506 may all be performed.

In step S503, when the recommended design is modified, the device 200 may generate a first learning signal indicating that the recommended design selected through the first input signal has been modified to another design.

For example, when the recommended design is set as the first design, if the first design is modified by replacing an image, changing the image placement position, etc., the device 200 sets the modified first design as the second design, and sets the first design as the second design. A first learning signal indicating that the design has been modified to a second design may be generated.

In step S504, the device 200 may input a first learning signal to the first artificial intelligence model and process the first artificial intelligence model to be learned and updated through the first learning signal.

That is, the learning device applies the first learning signal to the first artificial intelligence model to select the modified design as the recommended design when analyzing the recommended design of a company with an industry similar to that of the first company. Through the process of learning the artificial intelligence model, the first artificial intelligence model can be updated.

In step S505, when the recommended menu is modified, the device 200 may generate a second learning signal indicating that the recommended menu selected through the first input signal has been modified to another menu.

For example, when the recommended menu is set as the first menu and the first menu is modified by replacement, placement location change, etc., the device 200 sets the modified first menu as the second menu, and the first menu is changed to the second menu. A second learning signal indicating that the second menu has been modified may be generated.

In step S506, the device 200 may input a second learning signal to the second artificial intelligence model and process the second artificial intelligence model to be learned and updated through the second learning signal.

That is, the learning device applies the second learning signal to the second artificial intelligence model to select the modified menu as the recommended menu when analyzing the recommended menu of a company with a similar industry to that of the first company. Through the process of learning the artificial intelligence model, the second artificial intelligence model can be updated.

Figure 6 is a flowchart for explaining a process of learning modifications to menu contents according to an embodiment.

According to one embodiment, each step shown in FIG. 6 may be performed after step S410.

Referring to FIG. 6, first, in step S601, the device 200 may receive a request to modify the contents of the first page from the first user terminal 110. Here, a modification request is a request to modify the contents of the first page because there is a part to be modified in the contents of the first page.

In step S602, the device 200 may modify the content of the first page according to a request for modification of the content of the first page.

In step S603, when the content of the first page is modified, the device 200 may generate a third learning signal indicating that the recommended answer generated through the second input signal has been modified to different content.

For example, when the recommended answer is set as the first answer, if the first answer is modified by changing some content, the device 200 sets the modified first answer as the second answer, and the first answer is set as the second answer. A third learning signal may be generated to indicate that the answer has been modified.

In step S604, the device 200 may input a third learning signal to the third artificial intelligence model and process the third artificial intelligence model to be learned and updated through the third learning signal.

In other words, the learning device applies the third learning signal to the third artificial intelligence model to select the modified answer as the recommended answer when analyzing the recommended answer to be displayed as the contents of a menu similar to the first menu. Through the process of learning the intelligence model, the third artificial intelligence model can be updated.

Figure 7 is a flow chart to explain the process of selecting a recommended design according to an embodiment.

Referring to Figure 7, first, in step S701, when the first input signal is input, the first artificial intelligence model can confirm the industry of the first company as the first industry based on the first input signal.

In step S702, the first artificial intelligence model may classify the homepage whose design is set as the default design among pre-made homepages into the first group. For this purpose, among the homepages created for each industry stored in the database, the homepage design is set as the default design for some.

In step S703, the first artificial intelligence model may classify the homepage where the company's industry is set as the first industry among the homepages classified into the first group into the second group.

Specifically, the first artificial intelligence model classifies the homepages with the homepage design set as the default design among the homepages created for each industry stored in the database into the first group, and then selects the homepages among the homepages classified into the first group. Homepages where the operating company's industry is set to the first industry can be classified into the second group.

In step S704, the first artificial intelligence model may assign a higher first score to each homepage classified into the second group as the average number of daily visits to the homepage increases.

For example, if the second and third homepages are classified into the second group, the first artificial intelligence model determines that the average number of daily visits to the second homepage is 50, and assigns the first score to the second homepage. 5 points are given, and if the average number of daily visits to the third homepage is confirmed to be 60, the first score of 6 points can be given to the third homepage.

In step S705, the first artificial intelligence model may assign a higher second score to each homepage classified into the second group as the average access maintenance time of the homepage is longer.

For example, if the second homepage and the third homepage are classified into the second group, the first artificial intelligence model determines that the average access maintenance time of the second homepage is 5 minutes, and then determines the second score for the second homepage. 5 points are given, and if the average access maintenance time of the third homepage is confirmed to be 6 minutes, a second score of 6 points can be given to the third homepage.

In step S706, the first artificial intelligence model may assign a higher third score to each homepage classified into the second group as the most recent update time of the homepage is higher.

For example, if the second and third homepages are classified into the second group, the first artificial intelligence model determines that the update time of the second homepage was 5 days ago, and assigns a third score to the second homepage. 5 points are given, and if the update time of the third homepage is confirmed to be 4 days ago, a third score of 6 points can be given to the third homepage.

The first artificial intelligence model can set the maximum value that can be given to each of the first score, second score, and third score to the same score, and depending on importance, each of the first score, second score, and third score can be set to the same score. The maximum value that can be given can be set to different scores.

In step S707, the first artificial intelligence model may calculate a total score by adding up the first score, second score, and third score for each homepage classified into the second group.

In step S708, the first artificial intelligence model can compare the total scores of each homepage classified into the second group and identify the homepage with the highest total score among the homepages classified into the second group as the second homepage.

In step S709, the first artificial intelligence model may select a recommended design using the design of the second homepage.

In other words, if the homepage with the highest total score among the homepages classified into the second group is identified as the second homepage, the first artificial intelligence model can select the design of the second homepage as the recommended design.

In step S710, the first artificial intelligence model may generate a first output signal representing the selected recommended design.

Figure 8 is a flowchart for explaining the process of selecting a recommended menu according to an embodiment.

Referring to Figure 8, first, in step S801, when the first input signal is input, the second artificial intelligence model can confirm the industry of the first company as the first industry based on the first input signal.

In step S802, the second artificial intelligence model may classify the homepage whose menu is set as the default menu among the pre-made homepages into the third group. For this purpose, among the homepages created for each industry stored in the database, some have the homepage menu set as the default menu.

In step S803, the second artificial intelligence model may classify the homepage where the company's industry is set as the first industry among the homepages classified into the third group into the fourth group.

Specifically, the second artificial intelligence model classifies the homepages with the homepage menu set as the default menu among the homepages created for each industry stored in the database into the third group, and then selects the homepage among the homepages classified into the third group. Homepages where the operating company's industry is set as the first industry can be classified into the fourth group.

In step S804, the second artificial intelligence model may assign a higher fourth score to each homepage classified into the fourth group as the average number of daily visits to the homepage increases.

For example, if the second and third homepages are classified into the fourth group, the second artificial intelligence model determines that the average number of daily visits to the second homepage is 50, and assigns the fourth score to the second homepage. 5 points are given, and if the average number of daily visits to the third homepage is confirmed to be 60, a fourth score of 6 points can be given to the third homepage.

In step S805, the second artificial intelligence model may assign a higher fifth score to each homepage classified into the fourth group as the average access maintenance time of the homepage is longer.

For example, if the second and third homepages are classified into the fourth group, the second artificial intelligence model determines that the average access maintenance time of the second homepage is 5 minutes, and then gives the second homepage a 5th score. is given 5 points, and if the average access maintenance time of the third homepage is confirmed to be 6 minutes, a fifth score of 6 points can be given to the third homepage.

In step S806, the second artificial intelligence model may assign a higher sixth score to each homepage classified into the fourth group, the more recently the homepage was updated.

For example, if the 2nd and 3rd homepages are classified into the 4th group, the 2nd artificial intelligence model determines that the update time of the 2nd homepage was 5 days ago, and sets the 6th score for the 2nd homepage. 5 points are given, and if the update time of the third homepage is confirmed to be 4 days ago, a sixth score of 6 points can be given to the third homepage.

The second artificial intelligence model can set the maximum value that can be given to each of the fourth score, fifth score, and sixth score to the same score, and depending on importance, each of the fourth score, fifth score, and sixth score can be set to the same score. The maximum value that can be given can be set to different scores.

In step S807, the second artificial intelligence model may calculate the total score by adding up the fourth score, fifth score, and sixth score for each homepage classified into the fourth group.

In step S808, the second artificial intelligence model can compare the total scores of each homepage classified into the fourth group and identify the homepage with the highest total score among the homepages classified into the fourth group as the third homepage.

In step S809, the second artificial intelligence model can select a recommended menu using the menu of the third homepage.

In other words, the second artificial intelligence model can select the menu of the third homepage as the recommended menu if the homepage with the highest total score among the homepages classified into the fourth group is confirmed to be the third homepage.

In step S810, the second artificial intelligence model may generate a second output signal representing the selected recommended menu.

9 to 11 are flowcharts for explaining the process of setting the background color of the homepage according to one embodiment.

According to one embodiment, each step shown in FIGS. 9 to 11 may be performed after step S402.

Referring to FIG. 9 , first, in step S901, the device 200 may receive a first image, which is a logo image of a first company, from the first user terminal 110.

In step S902, the device 200 may divide one area of the first image into a first area, and divide the remaining area other than the first area into a second area.

For example, the device 200 may divide the left area of the first image into a first area and the right area of the first image as a second area.

In step S903, the device 200 analyzes the color displayed in the first area for each pixel, classifies all pixels in the first area according to color series, analyzes the color displayed in the second area for each pixel, and classifies the color displayed in the second area for each pixel. All pixels in area 2 can be classified according to color series. Here, the color series can be classified into one of predetermined color series such as red, orange, yellow, green, blue, indigo, and purple.

For example, a first pixel and a second pixel are included in the first area, the RGB value of the first pixel is (250, 10, 0), and the RGB value of the second pixel is (250, 0, 10). In this case, the device 200 may classify the first pixel and the second pixel into the red color series. That is, the device 200 can classify pixels that display a color somewhat similar to red into the red color series. At this time, if the first area consists of 500 pixels, the device 200 can analyze the color series for each of the 500 pixels and classify each of the 500 pixels according to the color series.

In step S904, the device 200 may confirm that the background color of the first image is the first color.

Specifically, the device 200 classifies all pixels of the first image according to the color series, and as a result of classifying all the pixels of the first image according to the color series, the color series with the largest number of pixels in the first image is the first image. If the color series is confirmed, the background color of the first image can be confirmed as the first color.

In step S905, if the background color of the first image is confirmed to be the first color, the device 200 sets the total number of pixels in the first area to the first number of pixels, and sets the number of pixels classified into the first color series in the first area. may be set as the second number of pixels, the total number of pixels in the second area may be set as the third number of pixels, and the number of pixels classified into the first color series in the second area may be set as the fourth number of pixels.

In step S906, the device 200 may calculate the first ratio by dividing the second number of pixels by the first number of pixels, and calculate the second ratio by dividing the fourth number of pixels by the third number of pixels.

In step S907, the device 200 may check whether the first rate is higher than the second rate.

If it is confirmed that the first ratio is higher than the second ratio in step S907, in step S1001, the device 200 selects the color series with the largest number of pixels from the color series other than the first color series in the first area as the second color series. You can check this.

In step S1002, the device 200 may set the number of pixels classified into the second color series in the first area as the fifth number of pixels.

In step S1003, the device 200 may calculate the third ratio as the value obtained by dividing the fifth number of pixels by the first number of pixels.

In step S1004, the device 200 may check whether the third ratio is higher than the reference ratio. Here, the reference ratio may be set differently depending on the embodiment.

If the third ratio is confirmed to be higher than the reference ratio in step S1004, the device 200 may set the second color as the background color of the first homepage in step S1005.

If it is confirmed that the third ratio is lower than the reference ratio in step S1004, in step S1006, the device 200 may set the number of pixels classified into the second color series in the second area as the sixth number of pixels.

In step S1007, the device 200 may calculate the fourth ratio as the value obtained by dividing the sixth number of pixels by the third number of pixels.

In step S1008, the device 200 may check whether the sum of the third ratio and the fourth ratio is higher than the reference ratio.

If the sum of the third ratio and the fourth ratio is confirmed to be higher than the standard ratio in step S1008, the device 200 may set the second color as the background color of the first homepage in step S1005.

If it is confirmed in step S1008 that the value added by the third ratio and the fourth ratio is lower than higher than the reference ratio, in step S1009, the device 200 selects the remaining colors excluding the first color series and the second color series in the first area. The color series with the largest number of pixels among the series can be identified as the third color series.

In step S1010, the device 200 may set a color that is a mixture of the second color and the third color as the background color of the first homepage.

Meanwhile, if it is confirmed in step S907 that the first ratio is lower than the second ratio, in step S1101, the device 200 selects the color series with the largest number of pixels from the color series other than the first color series in the second area. It can be identified as the fourth color series.

In step S1102, the device 200 may set the number of pixels classified into the fourth color series in the second area as the seventh number of pixels.

In step S1103, the device 200 may calculate the fifth ratio as the value obtained by dividing the seventh number of pixels by the third number of pixels.

In step S1104, the device 200 may check whether the fifth ratio is higher than the reference ratio.

If it is confirmed that the fifth ratio is higher than the reference ratio in step S1104, the device 200 may set the fourth color as the background color of the first homepage in step S1105.

If it is confirmed in step S1104 that the fifth ratio is lower than the reference ratio, in step S1106, the device 200 may set the number of pixels classified into the fourth color series in the first area as the eighth number of pixels.

In step S1107, the device 200 may calculate the sixth ratio as the value obtained by dividing the eighth number of pixels by the first number of pixels.

In step S1108, the device 200 may check whether the sum of the fifth ratio and the sixth ratio is higher than the reference ratio.

If the sum of the fifth ratio and the sixth ratio is confirmed to be higher than the standard ratio in step S1108, the device 200 may set the fourth color as the background color of the first homepage in step S1105.

If it is confirmed in step S1108 that the value added by the fifth ratio and the sixth ratio is lower than the reference ratio, in step S1109, the device 200 selects the remaining colors excluding the first color series and the fourth color series in the second area. Among the series, the color series with the largest number of pixels can be identified as the fifth color series.

In step S1110, the device 200 may set a color that is a mixture of the fourth color and the fifth color as the background color of the first homepage.

As described above, when the background color of the first homepage is set, the device 200 can create the first homepage by applying the background color of the first homepage.

Figure 12 is an exemplary diagram of the configuration of a device according to an embodiment.

Device 200 according to one embodiment includes a processor 210 and memory 220. The processor 210 may include at least one device described above with reference to FIGS. 1 to 11 or may perform at least one method described above with reference to FIGS. 1 to 11 . A person or organization using the device 200 may provide services related to some or all of the methods described above with reference to FIGS. 1 to 11 .

The memory 220 may store information related to the methods described above or store a program in which methods described later are implemented. Memory 220 may be volatile memory or non-volatile memory.

The processor 210 can execute programs and control the device 200. The code of the program executed by the processor 210 may be stored in the memory 220. The device 200 is connected to an external device (eg, a personal computer or a network) through an input/output device (not shown) and can exchange data through wired or wireless communication.

The device 200 may be used to learn an artificial intelligence model or use a learned artificial intelligence model. Memory 220 may include a learning or learned artificial intelligence model. The processor 210 may learn or execute the artificial intelligence model algorithm stored in the memory 220. The learning device that trains the artificial intelligence model and the device 200 that uses the learned artificial intelligence model may be the same or may be separate.

The embodiments described above may be implemented with hardware components, software components, and/or a combination of hardware components and software components. For example, the devices, methods, and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, and a field programmable gate (FPGA). It may be implemented using one or more general-purpose or special-purpose computers, such as an array, programmable logic unit (PLU), microprocessor, or any other device capable of executing and responding to instructions. A processing device may execute an operating system (OS) and one or more software applications that run on the operating system. Additionally, a processing device may access, store, manipulate, process, and generate data in response to the execution of software. For ease of understanding, a single processing device may be described as being used; however, those skilled in the art will understand that a processing device includes multiple processing elements and/or multiple types of processing elements. It can be seen that it may include. For example, a processing device may include a plurality of processors or one processor and one controller. Additionally, other processing configurations, such as parallel processors, are possible.

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

Software may include a computer program, code, instructions, or a combination of one or more of these, which may configure a processing unit to operate as desired, or may be processed independently or collectively. You can command the device. Software and/or data may be used on any type of machine, component, physical device, virtual equipment, computer storage medium or device to be interpreted by or to provide instructions or data to a processing device. , or may be permanently or temporarily embodied in a transmitted signal wave. Software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

Although the embodiments have been described with limited drawings as described above, those skilled in the art can apply various technical modifications and variations based on the above. For example, the described techniques are performed in a different order than the described method, and/or components of the described system, structure, device, circuit, etc. are combined or combined in a different form than the described method, or other components are used. Alternatively, appropriate results may be achieved even if substituted or substituted by an equivalent.

Therefore, other implementations, other embodiments, and equivalents of the claims also fall within the scope of the following claims.

Claims (3)

In a method of automating the creation of homepage production information using an artificial intelligence model performed by a device,
Receiving industry information indicating the industry of the first company from the first user terminal;
Generating a first input signal by encoding the industry information;
The first input signal is input to a first artificial intelligence model learned to recommend the design of the homepage according to the industry, and when the recommended design is selected through the first input signal, the recommended design is selected from the first artificial intelligence model. obtaining a first output signal representing;
The first input signal is input to a second artificial intelligence model trained to recommend menus of the homepage according to the industry, and when a recommended menu is selected through the first input signal, the recommended menu is selected from the second artificial intelligence model. obtaining a second output signal representing;
creating a first homepage based on the recommended design and the recommended menu;
transmitting production information of the first homepage to the first user terminal;
Upon receiving a confirmation request for the design and menu of the first homepage from the first user terminal, confirming the recommended design as the design of the first homepage and confirming the recommended menu as the menu of the first homepage ;
If it is confirmed that information about the first item is needed to create the contents of the first menu among the menus of the first homepage, based on the pre-registered first company information, extracting first item information, which is information;
Generating matching information by matching the first menu and the first item information;
Generating a second input signal by encoding the matching information;
The second input signal is input to a third artificial intelligence model learned to answer the contents of the menu according to the menu and item, and when a recommended answer is generated through the second input signal, the recommendation is made from the third artificial intelligence model. obtaining a third output signal representing an answer;
Creating a first page on which contents of the first menu are displayed based on the recommended answer; and
Comprising the step of transmitting production information of the first page to the first user terminal,
A method of automating the creation of homepage production information using an artificial intelligence model. delete According to paragraph 1,
After transmitting production information of the first homepage to the first user terminal,
Upon receiving a request for modification of the design and menu of the first homepage from the first user terminal, modifying at least one of the recommended design and the recommended menu according to the request for modification of the design and menu of the first homepage ;
When the recommended design is modified, a first learning signal is generated indicating that the recommended design selected through the first input signal has been modified to another design, and the first learning signal is input to the first artificial intelligence model. , processing the first artificial intelligence model to be learned and updated through the first learning signal; and
When the recommended menu is modified, a second learning signal is generated indicating that the recommended menu selected through the first input signal has been modified to another menu, and the second learning signal is input to the second artificial intelligence model. , further comprising processing the second artificial intelligence model to be learned and updated through the second learning signal,
After transmitting the production information of the first page to the first user terminal,
Upon receiving a request to modify the content of the first page from the first user terminal, modifying the content of the first page according to the request for modification of the content of the first page; and
When the content of the first page is modified, a third learning signal is generated indicating that the recommended answer generated through the second input signal has been modified to different content, and the third learning signal is transmitted to the third artificial intelligence model. Further comprising the step of processing the third artificial intelligence model to be learned and updated through the third learning signal by inputting,
A method of automating the creation of homepage production information using an artificial intelligence model.