KR20240001306A - Operation server and operation method for providing ai-based web builder by utilizing the change in the component's work and similar components - Google Patents

Abstract

An operation server and its operation method for providing an AI-based web builder by utilizing the work change amount of a component and similar components are disclosed. The operating server for providing the AI-based web builder provides a UI template selected by the user terminal to the user terminal, and provides a user interface that can place components to be placed within the UI template selected by the user terminal. Web Builder Service Management Department; a UI template management unit that receives and stores a plurality of UI templates in advance from an administrator, and generates and stores metadata for each of the stored UI templates; a component management unit that receives and stores a plurality of components in advance from an administrator, and generates and stores metadata for each of the stored components; and an artificial intelligence support unit that provides an artificial intelligence-based conversion support service to increase convenience in web design to the user terminal in parallel with the user interface.

Description

An operation server and its operation method for providing an AI-based web builder by utilizing the change in work of a component and similar components {OPERATION SERVER AND OPERATION METHOD FOR PROVIDING AI-BASED WEB BUILDER BY UTILIZING THE CHANGE IN THE COMPONENT'S WORK AND SIMILAR COMPONENTS}

The present invention relates to an operation server and its operating method for providing an AI-based web builder by utilizing the work change amount of a component and similar components.

MVP (Minimum viable product) is a product with minimum functionality and usually refers to a prototype application.

Developing such prototype applications generally requires high costs and time because it requires considerable IT manpower to be produced. However, most software startups often have difficulty hiring IT developers due to the nature of small and medium-sized businesses, and even if they outsource application development, they often have difficulty covering the outsourcing costs incurred during the development period.

In particular, the biggest reason for the failure of startup companies is the lack of market demand. Therefore, since it is necessary to quickly produce a prototype application and repeat market verification of the business item before entering the market, it is necessary to produce and modify the prototype application quickly and inexpensively.

In general, a web page or web-based application can be created by a person with relevant knowledge, or by purchasing an already created template site and replacing the skin of the purchased site.

However, recently, web builders provide web-based builders to consumers and allow consumers to create desired web-based applications simply by moving and applying the desired pre-made template using the user interface (UI) of the builder. The service is being launched.

For this reason, an alternative is being proposed that allows even ordinary people who do not know software development languages or do not have development skills to create desired applications. However, typical web builder services use pre-made templates and replace the text or design of the template with a designated design. This is a service that can only be produced by changing it to .

Therefore, there is a problem in that consumers have to select and design all the processes, so there are still many roles and requirements given to consumers, which is putting a psychological burden on consumers who want to use web builder services.

The purpose of the present invention to solve the above problems is to provide an operation server and a method of operating the same for providing an AI-based web builder by utilizing the work change amount of the component and similar components.

One aspect of the present invention for achieving the above object provides an operation server for providing an AI-based web builder.

The operating server for providing the AI-based web builder provides a UI template selected by the user terminal to the user terminal, and provides a user interface that can place components to be placed within the UI template selected by the user terminal. Web Builder Service Management Department; a UI template management unit that receives and stores a plurality of UI templates in advance from an administrator, and generates and stores metadata for each of the stored UI templates; a component management unit that receives and stores a plurality of components in advance from an administrator, and generates and stores metadata for each of the stored components; and an artificial intelligence support unit that provides an artificial intelligence-based conversion support service to increase convenience in web design to the user terminal in parallel with the user interface.

The artificial intelligence support unit monitors the change state of the UI template or the components through the user interface of the user terminal, and implements a UI template or component in anticipation of the change in the user terminal based on the monitored change state. is created in advance, and a preview area of the generated UI template or component is created and displayed on the user terminal to overlap the user interface.

The artificial intelligence support unit monitors change operations on the UI template selected by the user terminal, and provides a UI template as a preview using a first artificial neural network based on the degree of change in the current UI template monitored. It further includes a template conversion support unit that generates.

The artificial intelligence support unit monitors change operations on the components of the user terminal, and converts components to generate a component to be provided as a preview using a second artificial neural network based on the degree of change in the monitored current components. It further includes a support department.

The artificial intelligence support unit further includes a transformation simulation learning data generator that generates learning data for supervised learning of the first artificial neural network and the second artificial neural network in advance.

The conversion simulation learning data generator collects web design performance images, extracts a plurality of key frames from the collected web design performance images, identifies components and UI templates in the extracted key frames, and The keyframes are classified into keyframes in which identical UI templates are identified to create first groups, and the extracted keyframes are classified into keyframes in which identical components are identified to create second groups. 1 The work change amount of the UI template is extracted from each of the groups, the work change amount of the component is extracted from each of the second groups, and the work change amount of the extracted UI template is coupled with the similar UI template of the corresponding UI template to create the first Learning data is generated, and second learning data is generated by coupling the work change amount of the extracted component with a similar component of the corresponding component.

Another aspect of the present invention for achieving the above object provides a system for providing an AI-based web builder.

The system includes a user terminal provided with a web builder service; and an operating server that provides the web-based application creation tool to the user terminal.

The operation server includes a web builder service management unit that provides a UI template selected by the user terminal to the user terminal and provides a user interface for arranging components to be placed within the UI template selected by the user terminal; a UI template management unit that receives and stores a plurality of UI templates in advance from an administrator, and generates and stores metadata for each of the stored UI templates; a component management unit that receives and stores a plurality of components in advance from an administrator, and generates and stores metadata for each of the stored components; and an artificial intelligence support unit that provides an artificial intelligence-based conversion support service to increase convenience in web design to the user terminal in parallel with the user interface.

Another aspect of the present invention for achieving the above object provides a method of operating an operation server to provide an AI-based web builder.

The operating method includes receiving input from a user terminal the type and purpose of a web-based application to be produced through a web builder; providing a first user interface to the user terminal for selecting one of UI templates collected and stored in advance based on the input type and purpose; Displaying a UI template selected by the user terminal on the user terminal and providing a second user interface to the user terminal for arranging components to be placed within the selected UI template; and providing a third user interface to the user terminal capable of changing the design of the selected UI template displayed in the second user interface and components arranged in the selected UI template.

The operating method includes monitoring a change operation on the selected UI template and generating a UI template to be provided as a preview using a first artificial neural network based on the monitored degree of change of the current UI template; and monitoring change operations on the components of the user terminal and generating a component to be provided as a preview using a second artificial neural network based on the degree of change in the monitored current components.

Another aspect of the present invention for achieving the above object is to provide a non-transitory recording medium on which a program for executing the operating method of the operating server is recorded and can be read by a computer.

Another aspect of the present invention for achieving the above object provides a computer program recorded on a non-transitory recording medium in order to execute the operating method of the operating server in the operating server.

When using the operation server and its operation method to provide an AI-based web builder according to the present invention as described above, even if the consumer does not design and apply all elements directly, consumer convenience can be improved by utilizing AI-based design and components. We can provide an improved web builder.

Figure 1 is a schematic diagram showing an environment in which an operation server for providing an AI-based web builder according to an embodiment is performed.
Figure 2 is a diagram showing functional components of an operation server for providing an AI-based web builder according to an embodiment.
Figure 3 is a diagram for explaining the operation of a template conversion support unit according to an embodiment.
Figure 4 is a diagram for explaining the operation of the component conversion support unit according to one embodiment.
Figure 5 is a conceptual diagram for explaining the structure of a first artificial neural network and a second artificial neural network according to an embodiment.
Figure 6 is a representative flowchart showing the operation of the transformation simulation learning data generator according to an embodiment.
Figure 7 is a diagram illustrating the hardware configuration of an operation server for providing an AI-based web builder according to an embodiment.
Figure 8 is a diagram showing a wireless communication system that can be applied in a communication process according to an embodiment of the present invention.
FIG. 9 is a diagram showing a base station in the wireless communication system according to FIG. 8.
FIG. 10 is a diagram showing a terminal in the wireless communication system according to FIG. 8.
FIG. 11 is a diagram showing a communication interface in the wireless communication system according to FIG. 8.

Since the present invention can make various changes and have various embodiments, specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, and should be understood to include all changes, equivalents, and substitutes included in the spirit and technical scope of the present invention. While describing each drawing, similar reference numerals are used for similar components.

Terms such as first, second, A, and B may be used to describe various components, but the components should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another. For example, a first component may be named a second component, and similarly, the second component may also be named a first component without departing from the scope of the present invention. The term and/or includes any of a plurality of related stated items or a combination of a plurality of related stated items.

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

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

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the technical field to which the present invention pertains. 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.

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

Figure 1 is a schematic diagram showing an environment in which an operation server for providing an AI-based web builder according to an embodiment is performed.

Referring to Figure 1, the user terminal 200 of a user who wants to create a web-based application can access the operation server 100 for providing an AI-based web builder.

Web-based applications may include homepages, web applications, etc. that are accessed on the online web and run in web page format.

A web builder may refer to an online production tool for users to provide web-based applications using their user terminal 200. For example, the operation server 100 runs on the web to provide such a web builder, and may provide the web builder to the user terminal 200 connected to the operation server 100.

In one embodiment, the operation server 100 can collect and store a plurality of UI templates and a plurality of components in advance, and allows the user terminal 200 to select and place the collected UI templates and components. A user interface can be provided.

In one embodiment, UI templates are templates in the form of web pages designed to place pre-specified types of components in a pre-specified design structure, including an area where text can be entered and an image component can be placed. It may be a web page format with predefined areas, areas where menu bars can be placed, etc.

In one embodiment, components are various functional units that are pre-fabricated so that they can be placed within a UI template, and can perform pre-designated operations according to specific input data and visually display the results of the performed operations. For example, the components include a graph component that displays a graph corresponding to input data, a statistical output component that performs statistical operations based on the input data and displays result data according to the statistical operations, and performs image processing on the input data. and may include an image component that displays an image as a result of image processing. Here, pre-designated operations may be omitted in some cases and may be implemented to visually display the input data as is.

The operation server 100 inputs the type and purpose of the web-based application to be produced through the web builder to the user terminal 200, and provides UI templates to the user terminal 200 based on the input type and purpose. And the user terminal 200 can select one of the provided UI templates.

The operation server 100 displays the UI template selected by the user terminal 200 on the user terminal 200, and provides a user interface that can place components to be placed within the UI template displayed on the user terminal 200. It can be provided at (200).

The user of the user terminal 200 can select and place components desired to be placed within the UI template using the provided user interface.

The operation server 100 may provide the user terminal 200 with a user interface that can change the design of pre-collected components and UI templates. The user of the user terminal 200 can change the image, size, color, font, etc. that make up the components or UI templates using the provided user interface.

In addition, the operation server 100 may generate an access address for accessing the web-based application finally completed by the user terminal 200 and provide the generated access address to the user terminal 200.

The user of the user terminal 200 discloses the access address using another external site, etc., and the user terminal 300 of a third party that does not use the web builder uses the public access address to access the operating server through the access address. You can also access (100).

When a third party's user terminal 300 connects to the operation server 100 through the access address, the operation server 100 generates a web page for running a web-based application stored in advance by matching with the access address, A web-based application can be run within the created web page and provided to a third party's user terminal 300.

Through this, a third party's user terminal 300 can experience the web-based application created by the user's user terminal 200 in advance, and through this, the user of the user terminal 200 can experience the web-based application created by the user. Marketability tests can be performed to predict the market reaction in advance.

To this end, the operation server 100 stores various error details that occur in the process of using a web-based application run through the operation server 100 by a third party's user terminal 300 as a log file, and stores this in a log file. It can be provided to the user terminal 200 so that it can be viewed on the user terminal 200.

In one embodiment, the operation server 100 records the status of a third party's user terminal 300 using the web-based application (e.g., usage time, location of the user terminal 300 when used, user terminal 300 (mainly viewed data details, etc.) may be monitored, the monitored usage status details may be saved as a log file, and the log files may be provided to the user's user terminal 200 so that they can be viewed on the user's user terminal 200.

Figure 2 is a diagram showing functional components of an operation server for providing an AI-based web builder according to an embodiment.

Referring to FIG. 2, the operation server 100 for providing an AI-based web builder includes a web builder service management unit 101, a UI template management unit 102, a component management unit 103, and an artificial intelligence support unit 104. It can be included.

The web builder service management unit 101 can manage users of the user terminal 200. For example, the web builder service management unit 101 may receive user information from the user terminal 200, verify the received user information, and store the verified user information. Here, the user information includes not only personal information such as the name, address, ID, PASSWORD, and email address of the user of the user terminal 200, but also the type of application the user wants to create, the purpose of producing the application, and the production of the application. It may include a period, etc.

Optionally, the web builder service management unit 101 transmits payment information to the user terminal 200, performs payment in conjunction with an external financial server using the payment information transmitted by the user terminal 200, and makes the payment. A completion message may be received from an external financial server, and a user interface for the web builder service, a web-based production tool for creating web-based applications, may be provided to the user of the user terminal 200 for which payment has been completed.

The web builder service management unit 101 receives input from the user terminal 200 of the type and purpose of the web-based application to be produced through the web builder, and selects UI templates from among the UI templates collected and stored in advance based on the input type and purpose. A first user interface for selecting one is provided to the user terminal 200, and the user terminal 200 can select one of the UI templates using the first user interface.

The web builder service management unit 101 displays the UI template selected by the user terminal 200 on the user terminal 200 and places components to be placed within the UI template selected by the user terminal 200. A user interface may be provided to the user terminal 200.

The user of the user terminal 200 can select and place components desired to be placed within the UI template using the provided second user interface.

The web builder service management unit 101 may provide the user terminal 200 with a third user interface that can change the design of the UI template displayed in the second user interface and the components arranged in the corresponding UI template. The user of the user terminal 200 can change the image, size, color, font, etc. that make up the components or UI template using the provided third user interface.

In addition, the web builder service management unit 101 may generate an access address for accessing the web-based application finally completed by the user terminal 200 and provide the generated access address to the user terminal 200. .

The UI template management unit 102 may receive and store a plurality of UI templates in advance from the manager, and generate and store metadata for each of the stored UI templates. For example, when the source code of UI templates produced using external development personnel is uploaded to the operation server 100, the UI template management unit 102 reads the source code and generates metadata corresponding to the UI template. Then, a confirmation input for the generated metadata can be received from the administrator, and the metadata for which the confirmation input has been completed can be matched with the source code of the corresponding UI template and stored. For example, the UI template management unit 102 reads the source code of the UI template, collects the call functions constituting the source code, types of the collected call functions, parameters of the call functions, and calling order of the call functions. Metadata can be created based on etc.

In one embodiment, the metadata of the UI template is the theme type to which the corresponding UI template belongs (theme types can be input and specified in advance by the administrator, and the metadata is generated as data indicating one of the pre-designated meta types. may be) and the allowable conversion range of the components that make up the corresponding UI template. Here, the allowable conversion range may indicate the range of sizes, positions, and types in which the components constituting the corresponding UI template (e.g., font size, deployable size, etc.) can be changed. Taking font size as an example, the changeable size range of fonts placed at specific positions within the relevant UI template is included in the conversion allowable range. In a specific implementation, the conversion tolerance range has a maximum conversion tolerance range that can increase without changing other components when the size of the corresponding components increases, and when the size of the corresponding components decreases, the limit to which other components can grow is the maximum conversion tolerance range. It can be set to have the minimum conversion allowable range when the size difference between and exceeds a preset threshold. It may be desirable for the conversion tolerance range to be initially created by reading the source code and then finally set and input through the administrator's confirmation input.

In addition, the UI template management unit 102 calls the source code of the UI template selected by the user of the user terminal 200 from among the stored UI templates, and runs the called source code in the form of a web page on the web to display the user terminal ( 200).

The component management unit 103 may receive and store a number of components in advance from the manager, and generate and store metadata for each of the stored components. For example, when the source code of components produced using external development personnel is uploaded to the operation server 100, the component management unit 103 reads the source code and generates metadata corresponding to the component. Confirmation input for the metadata can be received from the administrator, and the confirmed metadata can be matched with the source code of the corresponding component and stored.

In one embodiment, the metadata of a component includes the type of function driven by the component (one of the types specified in advance by the administrator) and the conversion of the components (font, color, size, etc.) that make up the component. Allowable range may be included. Here, the conversion allowance range can be set in the same way as the conversion allowance range in the metadata of the previous UI template.

The artificial intelligence support unit 104 can provide an artificial intelligence-based conversion support service to increase convenience in web design to the user terminal 200 in parallel with the user interface provided by the web builder service management unit 101. .

For example, the artificial intelligence support unit 104 monitors the UI template or component change status through the third user interface of the user terminal 200, and changes the UI of the user terminal 200 based on the UI template or component change status. A UI template or component implemented by anticipating the template and component change form may be created in advance, and a preview area of the generated UI template or component may be created and displayed on the user terminal 200 to overlap the third user interface.

That is, the artificial intelligence support unit 104 continuously monitors the state in which the user of the user terminal 200 changes the UI template or component based on artificial intelligence, and prepares in advance the UI template or component that the user ultimately wants to convert and create. can be created.

To this end, the artificial intelligence support unit 104 monitors the UI template change operation of the user terminal 200 and previews it using the first artificial neural network 10 based on the degree of change in the monitored current UI template. It may include a template conversion support unit 104a that generates a UI template to be provided.

To the same purpose, the artificial intelligence support unit 104 monitors the component change operation of the user terminal 200 and provides a preview using the second artificial neural network 20 based on the degree of change of the current monitored component. It may include a component conversion support unit 104b that creates a component to be processed.

In addition, the artificial intelligence support unit 104 may further include a transformation simulation learning data generator 104c that generates learning data for supervised learning of the first artificial neural network 10 and the second artificial neural network 20 in advance. there is. The transformation simulation learning data generator 104c can generate learning data for supervised learning of the first artificial neural network 10 and the second artificial neural network 20 in advance based on various collected web design production images. .

Figure 3 is a diagram for explaining the operation of a template conversion support unit according to an embodiment.

Referring to FIG. 3, the template conversion support unit 104a acquires template metadata and template work monitoring data corresponding to the current UI template that the user of the user terminal 200 is currently working on, and stores the obtained template metadata. First input data can be generated using the template task monitoring data.

Specifically, the template conversion support unit 104a configures the theme type and conversion allowable range included in the template metadata, the image size variable included in the template work monitoring data, the image placement position variable, the image type variable, the component size variable, and the component placement. First input data including a positional variable and a component type variable may be generated. Here, the image size variable is the size variable (diagonal size change) when the user of the user terminal 200 manipulates the size of the image included in the UI template, and the image type variable is when the user changes the image included in the UI template to another image. When changed to , it means the type of image (for example, the object included in the image) after the change, and component size variation refers to the size variation (diagonal) when the user manipulates the size of a specific component included in the UI template. size change), the component placement position variable is the amount of position change (distance) when the user manipulates the position of a specific component included in the UI template, and the component type variable is the amount of change in position (distance) when the user manipulates the position of a specific component included in the UI template. When changed to a component, the component type (type) to which the changed component belongs can be indicated.

The form of the first input data is generally applied by constructing a matrix with each element as the component value. However, since it varies depending on the type of artificial neural network used and the number of nodes constituting the input layer, a person skilled in the art can determine the components. It can be generated from input data by adjusting appropriately.

When receiving first input data, the first artificial neural network 10 may output components for configuring a UI template to be provided as a preview as first output data. For example, the first artificial neural network 10 generates first output data including image size, image placement position, image type, component size, component placement position, and component format included in the UI template to be provided as a preview. can be output.

The template conversion support unit 104a may change the UI template that the user is currently working on based on the first output data and provide the changed UI template as a preview to the user terminal 200.

In one embodiment, the first artificial neural network 10 may be supervised in advance using training data.

Here, the learning data may be composed of a pair of a first training input value and a first training output value indicating the correct answer corresponding to the first training input value. The first training input values are image size variable, image placement position variable, image type variable, component size variable, component placement position variable, and component format variable obtained by monitoring the process of a person performing web design for a UI template. In addition, it may include the theme type and conversion allowable range that can be confirmed from the metadata of the corresponding UI template (or similar UI template).

The first training output value is the image size, image placement position, image type, component size, component placement position, and component that constitute the UI template (or similar UI template) finally created by a person through the web design process. It may be output data including a format.

The first artificial neural network 10 calculates a loss function using the output data obtained when the first training input value is input and the first output output value provided as the correct answer, and the result value of the calculated loss function This can be supervised to minimize this. Here, the process of supervised learning to minimize may be a process of readjusting the weights of the nodes constituting each layer of the first artificial neural network 10 so that the result of the loss function is minimized.

The first artificial neural network 10 and the second artificial neural network 20 are deep learning-based artificial neural networks, such as Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Autoencoder, Variational Autoencoder (VAE), Deep Residual Network (DRN), Generative Adversarial Network (GAN), etc. can be used, and it may be desirable to use artificial neural networks based on Recurrent Neural Network (RNN) or Long Short-Term Memory (LSTM). At this time, since various source codes for artificial neural networks based on Recurrent Neural Network (RNN) or Long Short-Term Memory (LSTM) are already public and accessible, those skilled in the art can evaluate the performance of one of these already known artificial neural networks. It is possible to select and utilize it. The present invention is not limited to a specific artificial neural network.

Figure 4 is a diagram for explaining the operation of the component conversion support unit according to one embodiment.

The component conversion support unit 104b, like the template conversion support unit 104a, acquires component metadata and component work monitoring data corresponding to the current component that the user of the user terminal 200 is currently working on, and configures the acquired component Second input data can be generated using metadata and component operation monitoring data.

Specifically, the component conversion support unit 104b includes the type of function in which the corresponding component included in the component metadata is classified, the conversion allowable range for which the corresponding component can be converted, the font type included in the component operation monitoring data, the component size variation, and second input data including the component color may be generated.

Here, the font type refers to the font type after changing when the user of the user terminal 200 changes the font type constituting the corresponding component, and the component size variation refers to the size variation when the user manipulates the size of the component ( diagonal size change), and component color variation refers to the color of the component after the change when the user changes the color of the component.

When receiving second input data, the second artificial neural network 20 may output components for configuring a component to be provided as a preview as first output data. For example, the second artificial neural network 20 may output second output data including the font type, component size, and component color that constitute the component to be provided as a preview.

The component conversion support unit 104b may change the UI component that the user is currently working on based on the second output data and provide the changed UI component as a preview to the user terminal 200.

In one embodiment, the second artificial neural network 20 may be supervised in advance using training data.

Here, the learning data may be composed of a pair of second training input values and a second training output value indicating the correct answer corresponding to the second training input value. The second training input value is the font type, component size variation, component color, and meta of the corresponding component (or similar component) obtained by monitoring the process of a person performing web design for the corresponding component or a similar component. It can include the type of function obtained through data and the allowable conversion range.

The second training output value may include the font type, component size, and component color that constitute the component finally created by a person through the web design execution process.

The second artificial neural network 20 calculates a loss function using the output data obtained when the second training input value is input and the second output output value provided as the correct answer, and the result value of the calculated loss function This can be supervised to minimize this. Here, the process of supervised learning to minimize may be a process of readjusting the weights of the nodes constituting each layer of the second artificial neural network 20 so that the result of the loss function is minimized.

The second artificial neural network 20 is a deep learning-based artificial neural network, including Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Autoencoder, Variational Autoencoder (VAE), Deep Residual Network (DRN), and Generative Adversarial. Network (GAN) can be used, and it may be desirable to use an artificial neural network based on Recurrent Neural Network (RNN) or Long Short-Term Memory (LSTM). At this time, since various source codes for artificial neural networks based on Recurrent Neural Network (RNN) or Long Short-Term Memory (LSTM) are already public and accessible, those skilled in the art can evaluate the performance of one of these already known artificial neural networks. It is possible to select and utilize it. The present invention is not limited to a specific artificial neural network.

Figure 5 is a conceptual diagram for explaining the structure of a first artificial neural network and a second artificial neural network according to an embodiment.

Although the structure of the first artificial neural network 10 is described with reference to FIG. 5, it should be interpreted that an artificial neural network with the same structure can be applied to the second artificial neural network 20 if the data size is different.

Referring to FIG. 5, the first artificial neural network 10 sequentially receives each column of the matrix constituting the first input data as an input vector (X), and calculates the number of component values of the input vector (X) and An input layer 11 composed of the same number (N) of input nodes, and a hidden layer 12 that transmits the output vector (Y') calculated using the output values received from the input layer 11 to the output layer 13. , and an activation function is applied to the output vector (Y`) to determine the probability (p) corresponding to the output vector (Y`), and the output layer ( 13) may be included. In the present invention, each of the nodes constituting the artificial neural network 10 may be interchangeably referred to as a neuron, which is an expression commonly used in the technical field to which the present invention belongs.

Specifically, when the artificial neural network 10 sequentially receives each column of the matrix provided as a training input as an input vector (X), the output vector (Y') obtained as the output of the hidden layer 12 and the training output value are vectorized. A loss function is calculated using the generated training output vector (Y), and supervised learning is performed to minimize the result of the calculated loss function.

For example, the loss function (H(Y,Y')) may be a cross entropy function. The cross entropy (H(Y,Y')) between the output vector (Y') and the training output vector (Y) can be defined as Equation 1 below.

In Equation 1, Ym may be the mth component of the training output vector (Y) (m is a natural number greater than or equal to 1), and Y`m may be the mth component of the output vector (Y`).

The input layer 11 can receive an input vector (X), apply one or more connection strength values corresponding to the input nodes to each component of the input vector (X), and transmit it to the hidden layer 12. there is.

For example, one or more connection strength values corresponding to each of the input nodes can be expressed as a first connection strength matrix (W _{N × M} ) having a size of N × M. At this time, N may be the same number as the input nodes, and M is set sufficiently small to be less than 1/10 times N. The first connection strength matrix (W _N×M ) may be a parameter that is set to an arbitrary initial value and then continuously updated through supervised learning.

In summary, the input layer 11 can transmit the intermediate operation vector (X) obtained by matrix multiplying the first connection strength matrix (W _{N × M} ) to the hidden layer 12.

The hidden layer 12 applies one or more connection strengths corresponding to each hidden node to the feature vector (F) obtained from the intermediate operation vector (X) received from the input layer 11 to obtain an output vector (Y `) can be generated, and the generated output vector (Y`) can be transmitted to the output layer 13.

For example, the hidden layer 12 may obtain a feature vector (F), which is a diagonal matrix that satisfies Equation 2 below, from the intermediate operation vector (X).

In Equation 2, R is a matrix that reversibly satisfies the relationship according to Equation 2 between the feature vector (F), which is a diagonal matrix, and the intermediate operation vector (X), and corresponds to a coordinate transformation matrix. The operation according to Equation 2 may be referred to as one of the diagonalization operations, and since it can be easily understood by those skilled in the art, detailed description will be omitted.

At this time, one or more connection strength values corresponding to each of the hidden nodes can be expressed as a second connection strength matrix (U _M×Q ) having a size of M×Q. That is, the second connection strength matrix (UM×Q) increases the feature vector (F) mapped into M dimensions back to Q dimensions. Q is set to a sufficiently large value, at least 10 times greater than M.

Meanwhile, the initial value of the second connection strength matrix (U _M×Q ) is set to a random value, and then the output generated by matrix multiplication between the feature vector (F) and the second connection strength matrix (U _M×Q ) The vector (Y') can be continuously updated to become the training output vector (Y), which is the training output value. That is, the second connection strength matrix (U _M×Q ) may also be a parameter that is updated as training data is continuously supervised.

In other words, the hidden layer 12 converts the intermediate operation vector ( The hidden layer 12 can be configured to perform a function similar to the convolutional layer responsible for feature extraction of a (convolutional neural network).

The output layer 13 determines the probability (p) corresponding to the output vector (Y') by applying an activation function to the output vector (Y') received from the hidden layer (12), and the determined probability (p) is the highest. An output vector (Y`) can be output. The activation function has the effect of converting values with various ranges into probabilities by expanding or reducing them to values between 0 and 1. For example, the activation function may be a ReLU function or a Softmax function, but is not limited thereto.

In one embodiment of the present invention, the hidden layer 12 may have the same number of neurons and may be composed of a plurality of hidden layers that are different from each other and connected sequentially. For example, the hidden layer 12 may be composed of a first hidden layer, a second hidden layer, a third hidden layer, and a fourth hidden layer that are different from each other and connected sequentially. At this time, the first to fourth hidden layers may be linear layers, and at least one of the first to fourth hidden layers may be a dense layer. Here, the designation of a linear layer or dense layer can be simply set through a predefined function in Tensorflow, so description is omitted.

Additionally, drop out may be applied to at least one of the first to fourth hidden layers according to a probability determined according to the number of hidden layers and the number of neurons each of the hidden layers has.

In one embodiment of the present invention, dropout refers to deactivating or disturbing some neurons by setting the weight of at least some neurons randomly selected according to probability to 0 or a predefined constant among the neurons constituting the corresponding layer. It may mean configuring it to do so.

For example, if the first hidden layer consists of 768 neurons and the probability (p) is 0.2, the The weight can be set to 0 or a predefined constant.

For example, probability (pp) can be defined as Equation 3 below.

In Equation 3, NRn may be the average number of neurons included in the first to fourth hidden layers, Nn may be the number of neurons included in the hidden layer to which dropout is to be applied, and HDn may be the total number of hidden layers, i.e. the first to fourth hidden layers. If it consists of fourth hidden layers, it may be 4.

In this way, when dropout is applied to at least one of the first to fourth hidden layers, the overfitting tendency that may occur as the number of hidden layers and the number of each neuron increases can be compensated or offset, resulting in a higher prediction effect. It can help you get it.

In one embodiment, among the first to fourth hidden layers, each of the first hidden layer and the third to fourth hidden layers may be composed of the same number of neurons. For example, each may consist of 768 neurons.

At this time, the number of neurons constituting the second hidden layer connected behind the first hidden layer may be at least four times larger than the number of neurons each of the other hidden layers has. If the number of neurons constituting the second hidden layer is set to be larger than the reference value than the number of neurons in each of the other hidden layers, there is an advantage in that the tendency of the artificial neural network 10 to overfit can be reduced.

Figure 6 is a representative flowchart showing the operation of the transformation simulation learning data generator according to an embodiment.

For supervised learning of the first artificial neural network (10) or the second artificial neural network (20), a vast amount of web design images of people working are required, and it is generally very difficult to obtain and apply them all. In particular, because web design work involves a wide variety of components and UI templates, it is virtually impossible to create learning data from work videos of the same components and UI templates as the components and UI templates you want to apply to the service.

In order to solve this problem, according to an embodiment of the present invention, components and UI templates are identified from a web design performance video, and then similar components and similar UI templates similar to the identified components and UI templates are collected in advance with the components and UI. We propose a method to compensate for the difficulty in collecting training data by selecting from a template and configuring training data.

Referring to FIG. 6, the conversion simulation learning data generator 104c can collect web design performance images (S100). Here, the web design performance video may be a video of a web designer or developer performing web design work on a component or a UI template including a component.

The transformation simulation learning data generator 104c can extract a plurality of key frames from the web design performance video (S110). Here, each keyframe may mean an image frame in which a component or UI template is identified.

For this purpose, a separate CNN-based artificial neural network can be constructed and used by supervised learning so that the constructed CNN-based artificial neural network can identify components or UI templates from the input image. CNN-based artificial neural networks are widely known as artificial neural networks that are easy to classify images, and can be trained to identify components and UI templates.

For example, a CNN-based artificial neural network can be easily obtained and used by a person skilled in the art with known artificial neural network structures such as LeNet, AlexNet, ZFNet, VGGNET, GoogLeNet, ResNet, and DensNet.

The transformation simulation learning data generator 104c can identify components and UI templates from the extracted keyframes (S120).

Specifically, in step S120, the transformation simulation learning data generator 104c may use the size of the image, the type of the image (i.e., the object contained in the image, and a known object identification algorithm) in the area identified by the UI template. ), first identification information including the size of the inserted component, the format of the component, and the function of the component can be identified.

Here, the component function can be identified by comparing the text included in the area identified as the component with texts stored in a pre-designated component function dictionary. The component function dictionary may be data storing phrases typically inserted into the component according to the function of the component.

In step S120, the conversion simulation learning data generator 104c may identify second identification information including the font type, size of the component, function of the component, and color of the component in the area identified as the component.

The transformation simulation learning data generator 104c classifies the extracted keyframes into keyframes in which identical UI templates are identified to create first groups, and divides the extracted keyframes into keyframes in which identical components are identified. Second groups can be created by classification (S130).

In step S130, the conversion simulation learning data generator 104c compares the first identification information between the identified UI templates, and when the difference between the first identification information is within a preset threshold difference, the conversion simulation learning data generator 104c compares the first identification information with the same UI template. You can decide.

In the same way, in step S130, the transformation simulation learning data generating unit 104c compares the second identification information between the identified components, and if the difference between the second identification information is within a preset threshold difference, the conversion simulation learning data generating unit 104c This can be decided by component.

In addition, in step S130, the transformation simulation learning data generator 104c may filter (remove) groups in which the number of key frames constituting each group is less than or equal to a preset threshold among the created first groups and second groups. there is. Through this process, it is possible to remove images that make it difficult to check the amount of change in web design work by excluding groups that do not have a sufficient number of image frames.

Next, the transformation simulation learning data generator 104c may extract the work change amount of the UI template from each of the first groups and the work change amount of the component from each of the second groups. Here, the work variation of the UI template includes image size variation, image placement position variation, image type variation, component size variation, component placement position variation, and component format variation for configuring the first input data.

In the same way, the work variation of the component includes the font type, component size variation, and component color for configuring the second input data.

The conversion simulation learning data generator 104c generates first learning data by coupling the work change amount of the extracted UI template with the similar UI template of the corresponding UI template, and converts the work change amount of the extracted component to the similar component of the corresponding component. Second learning data can be generated by coupling with .

Here, the first learning data is used for supervised learning of the first artificial neural network 10, and the second learning data is used for supervised learning of the second artificial neural network 20.

Here, the similar UI template calculates the template relatedness according to Equation 4 below in the relationship with the specific UI template identified in the web design performance video and each UI template collected and stored in advance, and the calculated template relatedness is It may be the smallest UI template among the saved UI templates.

TR is the template relevance, m and k are the large and small values of the total number of images inserted into each of the two UI templates being compared, and IS is included in the specific UI template identified in the web design performance video. is the size of the image, ISr is the size of the image included in the pre-saved UI template, EQ is a value indicating 1 if the image type is the same and 0 otherwise, and z2 and z1 are the two UI templates being compared, respectively. Among the total number of components inserted into each, the large and small values are, CPS is the size of the component inserted into a specific UI template, CPSr is the size of the component inserted into the pre-stored UI template, and EQn is the format of the component and If the component functions are the same, the value may be 1, and if they are different, the value may be 0. L1 and L2 are weights assigned to each image and component included in the UI template, and can be set according to the importance given to the administrator. The sum of L1 and L2 is 1, and L2 is set to a value that is at least twice as large as L1. It may be desirable to be

Here, the similar component calculates the component relatedness according to Equation 5 below from the relationship between the specific component identified in the web design performance video and the components collected and stored in advance, and the calculated component relatedness is selected among the components for which the component is pre-stored. It can be the smallest component.

T is a constant that has a preset first value (e.g., 0.8) if the font type of a specific component and the font type of a pre-stored component are the same and a preset second value (e.g., 1.2) if they are different, and CoPS is a specific This is the size of the component, CoPSr is the size of the pre-stored component, EF is a value indicating 1 if the functions of the two components being compared are the same and 0 if they are different, and sum(dr, dg, db) is the value of the two components being compared. The sizes of the components are resized to be the same (for example, resized to the smaller of the two), the color values that make up each of the two components are subdued between the color values at the same location, and the red color component (dr) of the subdued color values is It is the sum of both the green color component (dg) and the blue color component (db).

In this way, according to one embodiment of the present invention, even if the same web design images as the components and UI templates provided by the web builder are not collected, they can be collected and used as training data for the most similar components and UI templates according to predetermined conditions, It is easy to collect training data and easily establish the service.

After the service is launched, additional training is possible by collecting training data by directly monitoring the web design operation process of the user terminal 200 using the web builder. This allows training data to be used in the initial stage using similar data. Errors that occur can be sequentially corrected.

Figure 7 is a diagram illustrating the hardware configuration of an operation server for providing an AI-based web builder according to an embodiment.

Referring to FIG. 7, the operation server 100 for providing an AI-based web builder includes at least one processor 110, and instructions that instruct the at least one processor 110 to perform at least one operation ( It may include a memory 120 that stores instructions.

Here, the at least one processor 110 may mean a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor on which methods according to embodiments of the present invention are performed. You can.

The memory 120 may be comprised of at least one of a volatile storage medium and a non-volatile storage medium. For example, memory 120 may be one of read only memory (ROM) and random access memory (RAM),

The operation server 100 may further include a storage device 160 for storing input data, intermediate processing temporary data, output data, etc. for performing the at least one operation. For example, the storage device 160 may be a flash memory, a hard disk drive (HDD), a solid state drive (SSD), or various memory cards (eg, micro SD cards).

Additionally, the operation server 100 for providing an AI-based web builder may include a transceiver 130 that communicates through a wireless network. In addition, the operation server 100 for providing an AI-based web builder may further include an input interface device 140, an output interface device 150, a storage device 160, etc. Each component included in the operation server 100 for providing an AI-based web builder is connected by a bus 170 and can communicate with each other.

Examples of the operating server 100 for providing an AI-based web builder include a communication-capable desktop computer, a laptop computer, a laptop, a smart phone, and a tablet PC. PC, mobile phone, smart watch, smart glass, e-book reader, PMP (portable multimedia player), portable game console, navigation device, digital camera ), DMB (digital multimedia broadcasting) player, digital audio recorder, digital audio player, digital video recorder, digital video player, PDA (Personal Digital) Assistant), etc.

Figure 8 is a diagram showing a wireless communication system that can be applied in a communication process according to an embodiment of the present invention. FIG. 9 is a diagram showing a base station in the wireless communication system according to FIG. 8. FIG. 10 is a diagram showing a terminal in the wireless communication system according to FIG. 8. FIG. 11 is a diagram showing a communication interface in the wireless communication system according to FIG. 8.

Below, an example of a wireless communication network system supporting communication between the operation server 100 and the terminals 200 and 300 according to the present invention will be described in detail as an example. In the following description, the first node (device) may be an anchor/donor node or a centralized unit (CU) of an anchor/donor node, and the second node (device) may be a distributed unit (DU) of an anchor/donor node or relay node. It can be.

In a wireless communication system, a base station (BS), terminal, server, etc. may be included as part of the nodes that use a wireless channel.

A base station is a network infrastructure that provides wireless access to terminals and terminals. A base station has a coverage area defined as a geographic area depending on the distance over which signals can be transmitted.

A base station, like a "base station", can be referred to as an "access point (AP)", "enodeb (eNB)", "5th generation (5G) node", "wireless point", " It may be referred to as a “transmission/reception point (TRP)”.

Base stations, terminals, and terminals can transmit and receive wireless signals in the millimeter wave (mmWave) band (e.g., 28 GHz, 30 GHz, 38 GHz, 60 GHz). At this time, the base station, terminal, and terminal may perform beamforming to improve channel gain. Beamforming may include transmit beamforming and receive beamforming. That is, the base station, terminal, and terminal can give directivity to the transmitted signal and the received signal. To this end, the base station, terminal, and terminal can select a serving beam through a beam search procedure or beam management procedure. Thereafter, communication may be performed using a resource that is in a quasi co-located relationship with the resource carrying the serving beam.

A first antenna port and a second antenna port are considered to be quasi-colocated if the large-scale properties of the channel on which the symbols of the first antenna port are carried can be inferred from the channel on which the symbols of the second antenna port are carried. Large-scale properties may include one or more of delay spread, Doppler spread, Doppler shift, average gain, average delay, and spatial Rx parameters.

Below, a base station in the above-described wireless communication system is exemplified. As used hereinafter, the terms "-module", "-unit", or "-er" may mean a unit that processes at least one function or operation, and may mean hardware, software, or hardware and software. It can be implemented as a combination of.

The base station may include a wireless communication interface, a backhaul communication interface, a storage unit, and a controller.

The wireless communication interface performs the function of transmitting and receiving signals through a wireless channel. For example, a wireless communication interface may perform a conversion function between a baseband signal and a bit stream according to the system's physical layer standard. For example, in data transmission, a wireless communication interface encodes and modulates a transmitted bit stream to generate complex symbols. Additionally, when receiving data, the wireless communication interface demodulates and decodes the baseband signal to reconstruct the received bit stream.

The wireless communication interface performs the function of transmitting and receiving signals through a wireless channel. For example, a wireless communication interface may perform a conversion function between a baseband signal and a bit stream according to the system's physical layer standard. For example, in data transmission, a wireless communication interface encodes and modulates a transmitted bit stream to generate complex symbols. Additionally, upon receiving data, the wireless communication interface demodulates and decodes the baseband signal to reconstruct the received bit stream.

Additionally, the wireless communication interface up-converts the base band signal into an RF (Radio Frequency) band signal, transmits the converted signal through an antenna, and then down-converts the RF band signal received through the antenna into a base band signal. For this purpose, the wireless communication interface includes a transmission filter, a reception filter, an amplifier, a mixer, an oscillator, a digital-to-analog convertor (DAC), It may include an analog-to-digital convertor (ADC), etc. Additionally, the wireless communication interface may include a plurality of transmission and reception paths. Additionally, the wireless communication interface may include at least one antenna array including a plurality of antenna elements.

In terms of hardware, a wireless communication interface may include a digital unit and an analog unit, and the analog unit may include a plurality of subunits depending on operating power, operating frequency, etc. The digital unit may be implemented with at least one processor (eg, digital signal processor (DSP)).

The wireless communication interface transmits and receives signals as described above. Accordingly, a wireless communication interface may be referred to as a “transmitter,” “receiver,” or “transceiver.” Additionally, in the following description, transmission and reception performed through a wireless channel may be used to include processing performed in a wireless communication interface as described above.

The backhaul communication interface provides an interface to communicate with other nodes within the network. That is, the backhaul communication interface converts the bit stream transmitted to other nodes, for example, other access nodes, other base stations, upper nodes or the core network from the base station into physical signals, and converts the physical signals received from other nodes into bits. Convert to stream.

The storage unit stores data such as basic programs, applications, and setting information for operation of the base station. The storage unit may include volatile memory, non-volatile memory, or a combination of volatile memory and non-volatile memory.

The controller controls the overall operation of the base station. For example, the controller transmits and receives signals through a wireless communication interface or a backhaul communication interface. Additionally, the controller records data in the storage and reads the recorded data. The controller can perform the protocol stack functions required by communication standards. According to another implementation, the protocol stack may be included in the wireless communication interface. For this purpose, the controller may include at least one processor.

According to one embodiment, the controller may control the base station to perform operations according to the embodiment of the present invention.

According to various embodiments, a donor node of a wireless communication system includes at least one processor, a transceiver operably coupled to the at least one processor, and a plurality of radio bearers for a terminal accessing the relay node. configured to transmit to a relay node a first message containing first information related to the donor node regarding; receive a second message from the relay node containing second information related to the relay node regarding a plurality of radio bearers for the terminal; Data about the terminal can be transmitted to the relay node. Data may be transmitted to the terminal through a plurality of radio bearers based on the first information and the second information.

According to various embodiments, a radio bearer among a plurality of radio bearers may integrate a plurality of radio bearers. The at least one processor is further configured to determine a radio bearer for a terminal accessing the relay node and multiple radio bearers integrated by the radio bearer; Alternatively, a radio bearer for a terminal accessing the relay node can be determined.

According to various embodiments, the first message may include one or more of the following: identification of the terminal accessing the relay node; Display information indicating the type of terminal connecting to the relay node; Information about the radio bearer of the terminal connecting to the relay node; Information about the radio bearer carried by the terminal accessing the relay node; Information about the tunnel established for the radio bearer between the donor node and the relay node; Information on integrated multi-radio bearers; radio bearer mapping information; Information about the address on the donor node side; Information about the address on the relay node side; Display information corresponding to the radio bearer of the terminal connecting to the relay node; Indication information indicating the relay node to allocate a new address to a radio bearer for a terminal accessing the relay node; A list of address information that cannot be used by the relay node transmitting data of the radio bearer of the terminal connecting to the relay node; and information related to security configuration.

According to various embodiments, the second message may include one or more of the following: identification of the terminal accessing the relay node; Information about radio bearers approved by the relay node; Information about radio bearers not acknowledged by the relay node; Information about radio bearers partially accepted by the relay node; radio bearer mapping information; Configuration information of a terminal connecting to the relay node created by the relay node; Information about the address on the relay node side; and information related to security configuration.

According to various embodiments, the second message may further include information about integrated multiple radio bearers.

According to various embodiments, the donor node may include a central unit of the donor node, and the relay node may include a distributed unit of the donor node.

According to various embodiments, a relay node in a wireless communication system includes at least one processor, includes a transceiver operably coupled to the at least one processor, and transmits, from a donor node, a plurality of terminals accessing the relay node. configured to receive a first message containing first information related to a donor node regarding a radio bearer of; transmitting a second message containing second information related to a relay node regarding a plurality of radio bearers for the terminal to the donor node; Data about the terminal can be received from the donor node. Data may be transmitted to the terminal through a plurality of radio bearers based on the first information and the second information.

According to various embodiments, a radio bearer among a plurality of radio bearers may integrate a plurality of radio bearers. The at least one processor is further configured to determine a radio bearer for a terminal accessing the relay node and multiple radio bearers integrated by the radio bearer; Alternatively, multiple radio bearers integrated by radio bearer may be determined.

According to various embodiments, the first message may include one or more of the following: identification of the terminal accessing the relay node; Display information indicating the type of terminal connecting to the relay node; Information about the radio bearer of the terminal connecting to the relay node; Information about the radio bearer carried by the terminal accessing the relay node; Information about the tunnel established for the radio bearer between the donor node and the relay node; Information on integrated multi-radio bearers; radio bearer mapping information; Information about the address on the donor node side; Information about the address on the relay node side; Indication information corresponding to the radio bearer of the terminal connecting to the relay node; Indication information indicating the relay node to allocate a new address to a radio bearer for a terminal accessing the relay node; A list of address information that cannot be used by the relay node transmitting data of the radio bearer of the terminal connecting to the relay node; and information related to security configuration.

According to various embodiments, the second message may include one or more of the following: identification of the terminal accessing the relay node; Information about radio bearers approved by the relay node; Information about radio bearers not acknowledged by the relay node; Information about radio bearers partially accepted by the relay node; radio bearer mapping information; Configuration information of a terminal connecting to the relay node created by the relay node; Information about the address on the relay node side; and information related to security configuration.

According to various embodiments, the second message may further include information about integrated multiple radio bearers.

According to various embodiments, the donor node may include a central unit of the donor node, and the relay node may include a distributed unit of the donor node.

Below, the components of the terminal in the above-described wireless communication system are shown. The components of the terminal described below are components of a general-purpose terminal supported by a wireless communication system and may be merged or integrated with the components of the terminal according to the above-mentioned contents, and may be compared to the previous drawings to the extent of some overlap or conflict. The content explained with reference may be interpreted as having priority. The terms “-module”, “-unit”, or “-er” used below may refer to a unit that processes at least one function.

The terminal includes a communication interface, a storage unit, and a controller.

The communication interface performs the function of transmitting and receiving signals through a wireless channel. For example, the communication interface performs conversion between baseband signals and bit streams according to the system's physical layer standards. For example, in data transmission, a communication interface encodes and modulates the transmitted bit stream to generate complex symbols. Additionally, when receiving data, the communication interface demodulates and decodes the base band signal to reconstruct the received bit stream. Additionally, the communication interface up-converts the base band signal into an RF band signal, transmits the converted signal through an antenna, and then down-converts the RF band signal received through the antenna into a base band signal. For example, the communication interface includes a transmission filter, reception filter, amplifier, mixer, oscillator, and digital-to-analog convertor (DAC). , analog-to-digital convertor (ADC), etc.

Additionally, the communication interface may include multiple transmission and reception paths. Additionally, the communication interface may include at least one antenna array including a plurality of antenna elements. On the hardware side, the wireless communication interface may include digital circuits and analog circuits (eg, radio frequency integrated circuit, RFIC). A digital circuit may be implemented with at least one processor (e.g., DSP). The communication interface may include multiple RF chains. The communication interface may perform beamforming.

The communication interface transmits and receives signals as described above. Accordingly, a communication interface may be referred to as a “transmitter,” “receiver,” or “transceiver.” Additionally, in the following description, transmission and reception performed through a wireless channel may be used to include processing performed in the communication interface as described above.

The storage unit stores data such as basic programs, applications, and setting information for the operation of the terminal. The storage unit may include volatile memory, non-volatile memory, or a combination of volatile memory and non-volatile memory. Additionally, the storage unit provides stored data upon request from the controller.

The controller controls the overall operation of the terminal. For example, a controller sends and receives signals through a communication interface. Additionally, the controller records data in the storage and reads the recorded data. The controller can perform the protocol stack functions required by communication standards. According to another implementation, a protocol stack may be included in the communication interface. For this purpose, the controller may comprise at least one processor or microprocessor or may reproduce part of a processor. Additionally, a part of the communication interface or controller may be referred to as a communication processor (CP).

According to an embodiment of the present invention, a controller can control a terminal to perform an operation according to an embodiment of the present invention.

Below, a communication interface in a wireless communication system is illustrated.

The communication interface includes encoding and modulation circuitry, digital beamforming circuitry, a plurality of transmission paths, and analog beamforming circuitry.

Encoding and modulation circuitry performs channel encoding. At least one of a low-density parity check (LDPC) code, a convolutional code, and a polar code may be used for channel encoding. The encoding and modulation circuit generates modulation symbols by performing constellation mapping.

A digital beamforming circuit performs beam forming on digital signals (eg, modulation symbols). For this purpose, the digital beamforming circuit multiplexes the modulation symbols by beamforming weight values. Beamforming weights can be used to change the size and wording of the signal and may be referred to as a “precoding matrix” or “precoder.” The digital beamforming circuit outputs digital beamformed modulation symbols through a plurality of transmission paths. At this time, depending on the multiple input multiple output (MIMO) transmission method, modulation symbols may be multiplexed or the same modulation symbol may be provided to multiple transmission paths.

A plurality of transmission paths convert digital beamformed digital signals into analog signals. To this end, each of the plurality of transmission paths may include an inverse fast Fourier transform (IFFT) calculation unit, a cyclic prefix (CP) insertion unit, a DAC, and an upconversion unit. The CP insertion unit is for the orthogonal frequency division multiplexing (OFDM) method and can be omitted when applying another physical layer method (e.g., a filter bank multi-carrier (FBMC)). That is, multiple transmission paths provide independent signal processing processes for multiple streams generated through digital beamforming. However, depending on the implementation, some elements of multiple transmission paths may be commonly used.

The analog beamforming circuit performs beamforming on analog signals. For this purpose, the digital beamforming circuit multiplexes the analog signal by beamforming weighting values. Beamformed weights are used to change the size and wording of the signal. More specifically, depending on the connection structure between a plurality of transmission paths and antennas, the analog beamforming circuit can be configured in various ways. For example, each of the plurality of transmission paths may be connected to one antenna array. In another example, multiple transmission paths may be connected to one antenna array. In another example, multiple transmission paths may be adaptively connected to one antenna array or may be connected to two or more antenna arrays.

Methods according to the present invention may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. Computer-readable media may include program instructions, data files, data structures, etc., singly or in combination. Program instructions recorded on a computer-readable medium may be specially designed and constructed for the present invention or may be known and usable by those skilled in the computer software art.

Examples of computer-readable media may include hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, etc. Examples of program instructions may include machine language code such as that created by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. The above-described hardware device may be configured to operate with at least one software module to perform the operations of the present invention, and vice versa.

Additionally, the above-described method or device may be implemented by combining all or part of its components or functions, or may be implemented separately.

Although the present invention has been described above with reference to preferred embodiments, those skilled in the art may make various modifications and changes to the present invention without departing from the spirit and scope of the present invention as set forth in the claims below. You will understand that you can do it.

100: Operation server to provide AI-based web builder
200: user terminal
300: Third party user terminal

Claims (1)

As an operating server to provide an AI-based web builder,
A web builder service management unit that provides a UI template selected by a user terminal to the user terminal and provides a user interface for arranging components to be placed within the UI template selected by the user terminal;
a UI template management unit that receives and stores a plurality of UI templates in advance from an administrator, and generates and stores metadata for each of the stored UI templates;
a component management unit that receives and stores a plurality of components in advance from an administrator, and generates and stores metadata for each of the stored components; and
An artificial intelligence support unit that provides an artificial intelligence-based conversion support service to increase convenience in web design to the user terminal 200 in parallel with the user interface;
The artificial intelligence support department,
Monitoring the change status of the UI template or the components through the user interface of the user terminal, and generating and generating in advance a UI template or component implemented in anticipation of the change in the user terminal based on the monitored change status. Create a preview area of the UI template or component and display it on the user terminal to overlap the user interface,
The artificial intelligence support department,
A template conversion support unit that monitors change operations on the UI template selected by the user terminal and generates a UI template to be provided as a preview using a first artificial neural network based on the degree of change in the monitored current UI template. ; and
A component conversion support unit that monitors change operations on the components of the user terminal and generates a component to be provided as a preview using a second artificial neural network based on the degree of change in the current monitored components.
The artificial intelligence support department,
It further includes a transformation simulation learning data generator that generates learning data for supervised learning of the first artificial neural network and the second artificial neural network in advance,
The conversion simulation learning data generation unit,
Collect web design performance videos,
Extracting a plurality of key frames from the collected web design performance video,
Identify components and UI templates from the extracted keyframes,
Classifying the extracted keyframes into keyframes in which identical UI templates are identified to create first groups, and classifying the extracted keyframes into keyframes in which identical components are identified to create second groups; ,
Extracting the work change amount of the UI template from each of the first groups and extracting the work change amount of the component from each of the second groups,
First learning data is generated by coupling the work change amount of the extracted UI template with a similar UI template of the corresponding UI template, and the work change amount of the extracted component is coupled with the similar component of the corresponding component to generate second learning data. ,
An operating server to provide an AI-based web builder.