Method for generating web page on basis of website using generative artificial intelligence

Abstract

A method by which a server generates a web page on the basis of a website using generative artificial intelligence, according to the present invention, comprises the steps of: acquiring web information about the website; identifying a reference web page on the basis of the web information; partitioning the reference web page into a plurality of blocks by means of an artificial intelligence model; determining classification information on the basis of attribute information about the plurality of blocks; determining arrangement information about the plurality of blocks on the basis of the classification information; and providing editable content on the basis of the arrangement information about the plurality of blocks.

Description

A method for generating web pages based on a website using generative artificial intelligence.

[0001] The present invention relates to a method for generating a web page based on a website utilizing generative artificial intelligence and a server supporting the same.

[0002] With recent advancements in web technology, the need for websites is increasing, and the importance of creating customized web pages tailored to individual user needs is growing. Accordingly, automation tools are being actively developed across various industries to create web pages more quickly and efficiently.

[0003] However, existing tools often fail to accurately reflect users' specific needs or lack flexibility in layout and design. This increases the time and effort required by designers and developers, ultimately impacting the quality of web pages.

[0004] Therefore, there is a growing demand for automatic web page generation technology that can respond more accurately and quickly to users' specific needs.

[0005] The purpose of the present invention is to provide a method for a server to automatically generate a customized web page based on a reference web page through generative artificial intelligence.

[0006] In addition, the present invention aims to divide a reference web page into a plurality of blocks and provide editable content based on arrangement information for the blocks.

[0007] A method for generating a web page based on a website utilizing generative artificial intelligence of the present invention comprises a step of obtaining web information about the website, a step of confirming a reference web page based on the web information, a step of dividing the reference web page into a plurality of blocks using an artificial intelligence model, a step of determining classification information based on attribute information about the plurality of blocks, a step of determining arrangement information of the plurality of blocks based on the classification information, and a step of providing editable content based on the arrangement information about the plurality of blocks.

[0008] In one embodiment of the present invention, the step of determining the classification information may include the step of creating an outline for the plurality of blocks, the step of determining color information for each of the plurality of blocks based on the outline, and the step of determining the classification information based on the attribute information including the color information.

[0009] In one embodiment of the present invention, the step of generating the outline may include the step of determining the type of the main area based on texture pattern information corresponding to the main area, which is an area included in the outline, and, if the type is text, the step of identifying the text of the main area and generating the outline by extending it by a predetermined interval from the text.

[0010] In one embodiment of the present invention, the step of generating the outline may include, if the outlines overlap, a step of changing the overlapped outlines to include all blocks corresponding to the overlapped outlines.

[0011] In one embodiment of the present invention, the attribute information may be at least one of location information, direction information, structure information, content type information, and usage information of the block.

[0012] In one embodiment of the present invention, the step of determining the classification information may include a step of mapping the plurality of blocks to classification information included in a predetermined classification profile based on the attribute information.

[0013] In one embodiment of the present invention, the classification information may be candidate classification information having the highest corresponding degree of agreement among at least one candidate classification information.

[0014] In one embodiment of the present invention, the method may include a step of confirming request information for changing the classification information and a step of sequentially providing candidate classification information excluding the classification information based on the degree of agreement.

[0015] In one embodiment of the present invention, the step of determining the placement information may include a step of determining whether the classification information includes predetermined essential classification information, and a step of determining the placement information including the essential classification information based on the determination.

[0016] In one embodiment of the present invention, the step of providing the editable content may include the step of identifying content requiring replacement in the reference webpage through the artificial intelligence model, the step of identifying alternative content corresponding to the content requiring replacement, and the step of providing the editable content including the alternative content.

[0017] In one embodiment of the present invention, the step of checking the alternative content may include a step of receiving alternative content generated from the artificial intelligence model based on the content requiring replacement.

[0018] In one embodiment of the present invention, the replacement required content may include at least one of a trademark, a logo, an image, text, a graphic element, a video, and an audio clip.

[0019] The present invention has the advantage that a server can automatically generate a customized web page based on a reference web page through generative artificial intelligence.

[0020] In addition, the present invention has the advantage of being able to divide a reference web page into a plurality of blocks and provide editable content based on arrangement information for the blocks.

[0021] FIG. 1 is a diagram illustrating an example of a network environment according to one embodiment of the present invention.

[0022] FIG. 2 is a flowchart illustrating an example of a process for generating a web page based on a website utilizing generative artificial intelligence according to one embodiment of the present invention.

[0023] FIG. 3 is a flowchart illustrating an example of a process for determining classification information based on color information in a method for generating a web page based on a website utilizing generative artificial intelligence according to another embodiment of the present invention.

[0024] FIG. 4 is a flowchart illustrating an example of a process for changing content requiring replacement into alternative content in a method for generating a webpage based on a website utilizing generative artificial intelligence according to another embodiment of the present invention.

[0025] FIG. 5 is a diagram illustrating an example of a reference web page verified based on web information according to one embodiment of the present invention.

[0026] FIG. 6 is a drawing illustrating an example of generating outlines for a plurality of blocks according to one embodiment of the present invention.

[0027] FIG. 7 is a diagram illustrating an example of determining color information for each block based on an outline, according to one embodiment of the present invention.

[0028] FIG. 8 is a table illustrating an example of determining classification information based on classification information according to one embodiment of the present invention.

[0029] FIG. 9 is a diagram illustrating an example of providing editable content based on layout information according to one embodiment of the present invention.

[0030] Hereinafter, embodiments disclosed in this specification will be described in detail with reference to the attached drawings. Regardless of the drawing reference numerals, identical or similar components will be assigned the same reference numerals, and redundant descriptions thereof will be omitted. Furthermore, when describing embodiments disclosed in this specification, if a detailed description of a related known technology is judged to obscure the gist of the embodiments disclosed in this specification, the detailed description thereof will be omitted.

[0031] Terms that include ordinal numbers, such as first, second, etc., may be used to describe various components, but the components are not limited by these terms. These terms are used solely to distinguish one component from another.

[0032] Singular expressions include plural expressions unless the context clearly indicates otherwise.

[0033] In this application, each step described may be performed regardless of the listed order, except in cases where a special causal relationship requires that the steps be performed in the listed order.

[0034] In this application, terms such as “include” or “have” are intended to specify the presence of a feature, number, step, operation, component, part or combination thereof described in the specification, but should be understood not to exclude in advance the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts or combinations thereof.

[0035]

[0036] Hereinafter, the present invention will be described with reference to the attached drawings.

[0037] FIG. 1 is a diagram illustrating an example of a network environment according to one embodiment of the present invention.

[0038] A network environment according to an embodiment of the present invention illustrated in FIG. 1 may include a server (10) and an artificial intelligence model (20).

[0039] The server (10) is a device that provides a web page creation service that provides editable content based on a reference web page according to the user's requirements by utilizing an artificial intelligence model (20).

[0040] The network is not limited to a communication method, and can include a communication method that utilizes a communication network that the network can include (e.g., a mobile communication network, a wired Internet, a wireless Internet, a broadcasting network), and electrical signals (analog signals, digital signals, serial signals, Pulse Width Modulation), as well as short-range wireless communication.

[0041]

[0042] The server (10) according to various embodiments of the present invention may be implemented as a computer device or multiple computer devices that provide commands, codes, files, content, services, etc. The server (10) according to one embodiment may be an electronic device that can transmit and receive information by communicating with a receiving terminal that receives an artificial intelligence model (20) and editable content through a network. Although the server (10) is described as a single server for convenience of explanation, it may be configured as multiple servers, each of which provides different functions or services.

[0043] The server (10) may include a processor (11), a memory (12), and a communication unit (13). The processor (11) may control the overall operation of the memory (12) and the communication unit (13). According to various embodiments of the present invention, the memory (12) functions as a storage medium and may store a plurality of application programs running on the server (10), data, and commands for the operation of the server (10). In one embodiment, the memory (12) may be provided in the form of various storage devices such as ROM, RAM, flash drive, hard drive, etc. in terms of hardware, or may be provided in the form of web storage. The communication unit (13) may communicate with the artificial intelligence model (20) and the receiving terminal in a wired or wireless manner through a network.

[0044] The processor (11) can control the overall operation of the memory (12) and communication unit (13) to provide a web page generation service to the receiving terminal.

[0045] The memory (12) functions as a storage medium and can store a plurality of application programs running on the server (10), data for the operation of the server (10), and commands. In one embodiment, an application related to a web page creation service can be stored in the memory (12). This memory (12) can be provided in the form of various storage devices such as ROM, RAM, flash drive, hard drive, etc., in terms of hardware, or can be provided in the form of web storage.

[0046] The communication unit (13) can communicate with the artificial intelligence model (20) and the receiving terminal via a network in a wired / wireless manner.

[0047] The server (10) of the present invention performs a method of generating a web page based on a website utilizing generative artificial intelligence.

[0048] Specifically, the server (10) obtains web information about a website, verifies a reference web page based on the web information, divides the reference web page into a plurality of blocks through an artificial intelligence model, determines classification information based on attribute information for the plurality of blocks, determines arrangement information for the plurality of blocks based on the classification information, and provides editable content based on the arrangement information for the plurality of blocks.

[0049] Here, a website refers to a collection of related web pages accessible via the Internet. Specifically, web pages are created using web standard technologies such as HTML, CSS, and JavaScript, and can provide information or enable interaction for users.

[0050] Here, the reference webpage refers to a specific page of a website referenced by the webpage creation service. The server (10) analyzes the design, layout, content composition, and functionality of the reference webpage and utilizes this as reference material for creating editable content.

[0051] Here, web information refers to data or content collected from a website to verify a reference webpage. Specifically, it can include the address directly accessible to the reference webpage, screenshots or videos containing the visual layout and design of the reference webpage, and the entire code defining the page structure and style. For example, web information can include the website URL, webpage captures, and page source code.

[0052] Here, a block refers to a unit of content or functionality that is individually distinct within a webpage. Specifically, a block can contain web elements such as text, images, videos, forms, and buttons, and each block can have independent layout and style properties.

[0053] Here, attribute information refers to information about the characteristics used to analyze and classify blocks within a webpage. Specifically, attribute information may include the block's location, orientation, structure, content type, and purpose.

[0054] Here, classification information refers to data used to categorize blocks within a webpage according to specific categories or roles. Classification information is determined based on the block's attributes and defines each block's role in the layout based on its purpose, location, and function. For example, a block containing an image may be classified as a "main banner," while a block primarily containing text may be classified as a "header" or "body."

[0055] Here, layout information refers to information about how blocks within a webpage are positioned and arranged. Specifically, layout information includes the position, alignment, spacing, and interaction of each block, and can be used to determine the overall layout of a webpage.

[0056] Here, editable content refers to a webpage template that users can modify or adjust. Specifically, editable content can include various web elements, such as text, images, videos, links, and forms, and users can modify these elements to suit their needs and goals.

[0057]

[0058] The artificial intelligence model (20) may include a processor (21), a memory (22), and a communication unit (23). Each detailed component included in the artificial intelligence model (20) performs substantially the same function as each detailed component included in the server (10).

[0059] The artificial intelligence model (20) can analyze web information, classify and segment reference web pages, create editable web page templates, and check for content that needs to be replaced in the reference web page when the server (10) performs a web page creation service.

[0060] An artificial intelligence model (20) is a form of artificial intelligence used to generate or predict new data using generative AI based on existing data. Specifically, it can be a model that utilizes machine learning algorithms and deep learning networks to learn from input data and generate new data based on that learning.

[0061]

[0062] FIG. 2 is a flowchart illustrating an example of a process for generating a web page based on a website utilizing generative artificial intelligence according to one embodiment of the present invention.

[0063] In step (210), the server (10) obtains web information about the website.

[0064] Here, web information refers to data or content collected from a website to verify a reference webpage. Specifically, web information may be in a form accessible and provided by users of a webpage creation service. For example, a user provides the address of a desired reference webpage via a webpage's Uniform Resource Locator (URL), allowing the server (10) to access and analyze the page. Additionally, screenshots, videos, GIF captures, PDF files, or the source code of the reference webpage, which illustrate a specific design or layout of the webpage, may also be provided as web information.

[0065]

[0066] In step (220), the server (10) checks the reference web page based on web information.

[0067] Specifically, the server (10) can identify reference web pages to be used as models or examples in a web page generation service based on web information. Furthermore, the server (10) can analyze information such as provided URLs, screenshots, and source code to determine the design, layout, and functional characteristics of the reference web page.

[0068]

[0069] In step (230), the server (10) divides the reference web page into multiple blocks through an artificial intelligence model (20).

[0070] Here, a block refers to a unit of content or functionality that is individually distinguished within a webpage.

[0071] Specifically, the server (10) can scan a reference webpage using an artificial intelligence model (20) to identify blocks based on visual and structural elements. For example, chunks of text, groups of images, interactive elements, etc. can be classified into individual blocks, and each block can be independently identified.

[0072]

[0073] In step (240), the server (10) determines classification information based on attribute information for multiple blocks.

[0074] Specifically, the server (10) determines classification information according to the function and role of each block based on attribute information such as the location, size, content, and style of each block. For example, a wide, thin block located at the top of the page and containing a logo and navigation menu may be classified as a "header." A block containing promotional text along with a large image in the center of the page may be identified as a "main banner." Additionally, a block in which multiple product images and descriptions are arranged in a grid or list format may be classified as a "product gallery."

[0075] In determining classification information, the server (10) can map a plurality of blocks to classification information included in a predetermined classification profile based on attribute information.

[0076] Specifically, the server (10) can map the attribute information of each block according to a pre-stored classification system. For example, in the classification profile, headers can be classified into "left logo header" and "center logo header" based on their placement, and banners can be classified into "simple banner," "image-text," and "text-image" based on the arrangement order of the images and text.

[0077]

[0078] The classification information to be determined is the candidate classification information with the highest matching degree among at least one candidate classification information, and the server (10) can check request information for classification information change and sequentially provide candidate classification information based on the matching degree.

[0079] Specifically, the server (10) can evaluate several possible classifications for each block, score the suitability of each classification, and determine the classification with the highest score. For example, when the server (10) analyzes a specific section of a web page, it can evaluate whether the section belongs to a "header," "footer," or "main content," and if the user is not satisfied with this initial classification, the server (10) can provide the classification with the next highest match and allow the user to select it.

[0080]

[0081] In step (250), the server (10) determines the arrangement information of multiple blocks based on the classification information.

[0082] Specifically, the server (10) can determine layout information regarding the detailed layout arrangement of a web page template based on the classification information of a plurality of blocks. For example, if the block classification information of a reference web page is identified as a header, banner, three-column block, and gallery, the server (10) can determine layout information that organizes the web page in the order of header, banner, three-column block, and gallery from top to bottom. In addition, if a block essential to the classification information is missing, for example, if there is no header or footer, the server (10) can reconstruct the layout information by adding a header or footer as needed.

[0083] Alternatively, the server (10) may change the layout according to priority when determining layout information. For example, on a reference web page, even if the banner is located below the header, it may be placed in the first visible position when the page loads.

[0084]

[0085] In step (260), the server (10) provides editable content based on layout information for a plurality of blocks.

[0086] Specifically, the server (10) can set areas within the webpage template that users can easily modify based on the layout information of each block. For example, the header block can be configured to allow users to freely change the text of logos and menu items, and the main banner block can provide editing tools to allow users to directly update the headline and description displayed with the image. The product gallery block can be configured to allow users to add or remove each product image and description. Additionally, the server (10) can provide editable content that responds appropriately to the various screen sizes and devices of the receiving terminal.

[0087]

[0088] FIG. 3 is a flowchart illustrating an example of a process for determining classification information based on color information in a method for generating a web page based on a website utilizing generative artificial intelligence according to another embodiment of the present invention.

[0089] In step (310), the server (10) generates outlines for multiple blocks.

[0090] Specifically, the server (10) can draw an outline using image analysis techniques such as edge detection, contour detection, and bounding boxes to define the visual boundaries of each block. For example, when edge detection is used, the server (10) can identify the boundaries of a block by finding points where the brightness of the image changes sharply, and thereby generate an outline of a complex-shaped object. In the case of contour detection technology, the server (10) can connect consecutive points in a block to generate a complete outline of an object. In addition, when a bounding box is used, the server (10) can recognize the location and size of a block and generate a minimal rectangular outline around it.

[0091]

[0092] In step (311), the server (10) determines the type of the main area based on texture pattern information corresponding to the main area, which is the area included in the outline.

[0093] Specifically, the server (10) can analyze the texture characteristics of each key area to identify its type. For example, an area with a consistent texture and pattern is generally considered a background, while an area with an irregular pattern and a variety of color combinations may be considered an image or graphic element. Additionally, an area exhibiting a consistent spacing and repeating letter pattern may be considered text.

[0094]

[0095] In step (313), if the type is text, the server (10) identifies the text of the main area and creates an outline by extending the text by a predetermined distance.

[0096] Specifically, the server (10) can trace the character boundaries of the text and add a certain amount of space to the end of each line of the text to form an extended outline encompassing the entire text block. For example, to clearly distinguish the area containing the text, the server (10) can allocate additional space to the top, bottom, left, and right of the text, thereby emphasizing the entire text block and clearly distinguishing it from adjacent images or other elements.

[0097]

[0098] In step (315), if the outlines overlap, the server (10) changes the overlapped outlines to include all blocks corresponding to the overlapped outlines.

[0099] Specifically, the server (10) can detect areas where the outlines of multiple generated blocks overlap each other and reconstruct this into a single, integrated outline, thereby enabling each block to be clearly distinguished. For example, if the outlines of a text block and an adjacent image block overlap each other, the server (10) can generate a new outline encompassing the corresponding text block and image block, thereby allowing the two elements to be processed as a single, interrelated group.

[0100]

[0101] In step (320), the server (10) determines color information for each of a plurality of blocks based on the outline.

[0102] Specifically, the server (10) can create an outline for each block, fill the outline with a color, and then analyze and extract the color of the area beyond the outline. For example, a primary color can be applied within the outline, and the area outside the outline can be processed differently to visually distinguish each block.

[0103]

[0104] In step (330), the server (10) determines classification information based on attribute information including color information.

[0105] Specifically, the server (10) can clearly distinguish each block through the main color applied within the outline, and then determine classification information for the distinguished blocks.

[0106]

[0107] FIG. 4 is a flowchart illustrating an example of a process for changing content requiring replacement into alternative content in a method for generating a webpage based on a website utilizing generative artificial intelligence according to another embodiment of the present invention.

[0108] In step (410), the server (10) checks for content requiring replacement on the reference webpage through an artificial intelligence model (20).

[0109] Specifically, the server (10) can specifically check trademarks, logos, product images containing trademarks, etc. For example, the server (10) can check whether a trademark or logo is included in product packaging images, advertising banner images, product usage images, etc.

[0110] The server (10) can input criteria for identifying content in need of replacement from the user and provide them to the artificial intelligence model (20) via a prompt. For example, the artificial intelligence model (20) can be instructed via a prompt to indicate that the 'Nike' logo and the 'Just Do It' slogan displayed on the reference webpage are copyrighted content, and to detect content containing similar visual elements or text. In addition, the user can determine the scope of the target for identifying content in need of replacement via the prompt. For example, the user can instruct the artificial intelligence model (20) via the prompt to check whether the 'Nike' logo and the 'Just Do It' slogan are included only in images excluding text on the webpage.

[0111] Additionally, the server (10) can check for content that requires replacement through the editing history of other users who created editable content based on the same reference web page.

[0112] Specifically, the server (10) can analyze the edit history of other users who have used the same reference webpage. Based on the edit history, the server (10) can identify patterns of frequently replaced content and use this to identify content in need of replacement. For example, if a specific logo or image has been replaced by multiple users, the server (10) can automatically detect the logo or image and suggest it as a replacement when a new user edits the webpage.

[0113]

[0114] In step (420), the server (10) checks for replacement content corresponding to the content requiring replacement.

[0115] Specifically, the server (10) can replace copyrighted elements with similar, legally sound content. For example, if a logo of a certain famous brand used on a reference webpage is copyrighted, the server (10) can replace the logo with a generic, non-symbolic icon or graphic. For copyrighted images, the server can replace them with images with similar, free licenses, or use an artificial intelligence model (20) to generate new images with similar themes. Alternatively, if text is copyrighted, the server can replace the text with new text that conveys the same information but is expressed differently.

[0116] The server (10) can receive alternative content generated by the artificial intelligence model (20) based on the content in need of replacement. Specifically, the artificial intelligence model (20) can generate alternative content that is similar to a copyrighted element but does not pose legal issues. For example, if a specific brand logo is displayed on a reference webpage, the artificial intelligence model (20) can generate a non-graphical representation of the logo that is visually similar but does not pose legal issues. Furthermore, in the case of copyrighted text, the artificial intelligence model (20) can generate new text that conveys the same information but is expressed differently and provide it to the server (10).

[0117]

[0118] In step (430), the server (10) provides editable content including alternative content.

[0119] Specifically, the server (10) may provide editable content with replacement content inserted in the location of the content requiring replacement, along with tools that allow the user to further adjust the content as desired. For example, if a copyrighted image is replaced, the new image is inserted in the original location, and the user can adjust the image's size, location, filter, etc., and replace it with a different image as needed. Furthermore, if text is changed, the new text is placed in the original location, and the user can freely adjust the font size or color or replace it with completely different content.

[0120]

[0121] Hereinafter, with reference to FIGS. 5 to 9, an example of a method for generating a web page based on a website utilizing generative artificial intelligence will be described.

[0122] FIG. 5 is a diagram illustrating an example of a reference web page verified based on web information according to one embodiment of the present invention.

[0123] Referring to FIG. 5, the reference webpage is an advertising webpage for a soccer shoe brand called 'GoldKick' and includes a header (510), a banner (520), a gallery (530), and product images (540). The header (510) is located at the top of the page, with the 'GoldKick' logo on the left, and a navigation menu in the center including brand introduction, shopping mall products, shopping mall guide, notices, frequently asked questions, etc. The banner (520) has a background image of running on a soccer field and the advertising phrase 'Unleash Speed, Unleash Power' displayed in the center. The gallery (530) consists of three images each introducing new products, MD recommended products, and bestsellers, and the product images (540) have four product images arranged horizontally to show the diversity of the products.

[0124]

[0125] FIG. 6 is a drawing illustrating an example of generating outlines for a plurality of blocks according to one embodiment of the present invention.

[0126] Referring to FIG. 6, a header (610), a banner (620), a gallery (630), and a product image (640) are displayed, each of which has an outline created to clearly define a visual boundary for the header (510), the banner (520), the gallery (530), and the product image (540) mentioned in FIG. 5.

[0127] In FIG. 6, the header (610) is formed by the outline of the entire header, including the 'GoldKick' logo on the left and the navigation menu in the center, and the gallery (630) is divided into three separate outlines, including each image and its corresponding description text. The product image (640) is displayed with an outline along with associated text for each of the four product images.

[0128] Additionally, the product text (645) traces the character boundaries of each text block and adds a certain margin to the end of each line, thereby creating an extended rectangular outline encompassing the entire text block.

[0129]

[0130] FIG. 7 is a diagram illustrating an example of determining color information for each block based on an outline, according to one embodiment of the present invention.

[0131] Referring to FIG. 7, the server (10) can fill in the outlines of the header (610), banner (620), gallery (630), product image (640), and product text (645) generated in FIG. 6 with colors, and then analyze and extract the colors of the areas beyond the outlines. Specifically, the server (10) can apply a primary color to the interior of the outlines to emphasize the visual characteristics of each block, and process the areas outside the outlines differently so that they can be visually distinguished, thereby determining classification information for the distinguished blocks.

[0132]

[0133] FIG. 8 is a table illustrating an example of determining classification information based on attribute information according to one embodiment of the present invention.

[0134] Referring to Figure 8, the attribute information of blocks A, B, C, D, and E, such as location, direction, structure, and content type information, is displayed, and the classification information of each block is determined based on the attribute information.

[0135] Referring to Figure 8, Block A is positioned horizontally at the top of the webpage, with a logo placed on the left side to immediately identify the website's brand. Based on this position and structural characteristics, the server (10) can determine the classification information for Block A as a "left logo header."

[0136] Referring to Figure 8, Block B is located at the top center of the webpage and features a structure combining a large image and promotional text. This serves to attract visitors' attention and emphasize the website's core message or event, so the server (10) can determine the classification information for Block B as "main banner."

[0137] Referring to Figure 8, Block D is located at the bottom of the webpage and combines images and text to form a three-tier, two-layer list structure. This structure, which includes descriptions and additional information for each image, effectively provides visitors with a variety of product information, and thus, the server (10) can determine the classification information of Block D as "Gallery."

[0138]

[0139] FIG. 9 is a diagram illustrating an example of providing editable content based on layout information according to one embodiment of the present invention.

[0140] Referring to FIG. 9, the editable content includes a header (910), a logo (915), a banner (920), a gallery (930), and a product image (940).

[0141] Referring to Fig. 9, editable content is generated based on the arrangement information of the blocks recognized in Fig. 7, and an editing tool is provided so that the user can easily modify the content within the webpage. The server (10) can identify text and product images containing the trademark and logo of 'GoldKick', which are copyrighted elements, in the reference webpage, and replace them with similar content that is not problematic in the legal sense. Referring to Fig. 9, the logo (915) within the header (910) is displayed by changing the logo and trademark of 'GoldKick' to a soccer ball image and the phrase 'SPORTS'. In addition, the product image (940) is displayed by changing the existing product image of 'GoldKick' to a general-purpose shoe icon.

[0142]

[0143] The technical features disclosed in each embodiment of the present invention are not limited to that embodiment, and, unless they are mutually incompatible, the technical features disclosed in each embodiment may be combined and applied to different embodiments.

[0144] Therefore, although each embodiment focuses on its own technical features, each technical feature can be applied in combination with each other as long as they are not mutually incompatible.

[0145] The present invention is not limited to the above-described embodiments and the attached drawings, and various modifications and variations are possible within the scope of those skilled in the art. Therefore, the scope of the present invention should be defined not only by the claims of this specification but also by equivalents thereof.

Claims

1. In a method for generating a web page based on a website using generative artificial intelligence, A step of obtaining web information about the above website; A step of checking a reference web page based on the above web information; A step of dividing the above reference web page into multiple blocks using an artificial intelligence model; A step of determining classification information based on attribute information for the plurality of blocks; A step of determining arrangement information of the plurality of blocks based on the above classification information; and A step of providing editable content based on the arrangement information for the plurality of blocks. A method for generating web pages based on a website using generative artificial intelligence.

2. In paragraph 1, The step of determining the above classification information is: A step of generating an outline for the above plurality of blocks; A step of determining color information for each of the plurality of blocks based on the outline; and A step of determining the classification information based on the attribute information including the color information. A method for generating web pages based on a website using generative artificial intelligence.

3. In paragraph 2, The steps for creating the above outline are: A step of determining the type of the main area based on texture pattern information corresponding to the main area, which is an area included in the above outline; and If the above type is text, the step of identifying the text of the main area and generating the outline by extending the text by a predetermined interval A method for generating web pages based on a website using generative artificial intelligence.

4. In paragraph 2, The steps for creating the above outline are: In the case where the above outlines overlap, a step of changing the overlapped outlines to include all blocks corresponding to the overlapped outlines is included. A method for generating web pages based on a website using generative artificial intelligence.

5. In paragraph 1, The above attribute information is at least one of the location information, direction information, structure information, content type information, and purpose information of the block. A method for generating web pages based on a website using generative artificial intelligence.

6. In paragraph 1, The step of determining the above classification information is: A step of mapping the plurality of blocks to classification information included in a predetermined classification profile based on the above attribute information. A method for generating web pages based on a website using generative artificial intelligence.

7. In paragraph 1, The above classification information is the candidate classification information with the highest corresponding degree of agreement among at least one candidate classification information. A method for generating web pages based on a website using generative artificial intelligence.

8. In paragraph 8, A step of confirming request information for changing the above classification information; and A step of sequentially providing the candidate classification information excluding the above classification information based on the above consistency. A method for generating web pages based on a website using generative artificial intelligence.

9. In paragraph 1, The step of determining the above placement information is: A step of determining whether the above classification information includes pre-determined essential classification information; and Based on the above judgment, a step of determining the placement information including the required classification information is included. A method for generating web pages based on a website using generative artificial intelligence.

10. In paragraph 1, The step of providing the above editable content is: A step of identifying content that needs to be replaced on the reference webpage using the artificial intelligence model; A step of checking alternative content corresponding to the above replacement required content; and comprising a step of providing the editable content including the alternative content; A method for generating web pages based on a website using generative artificial intelligence.

11. In paragraph 10, The steps to check the above alternative content are: A step of receiving alternative content generated from the artificial intelligence model based on the above alternative required content. A method for generating web pages based on a website using generative artificial intelligence.

12. In paragraph 10, The above replacement required content includes at least one of trademarks, logos, images, text, graphic elements, videos and audio clips. A method for generating web pages based on a website using generative artificial intelligence.

Images