Website creation support device

Abstract

To provide a web site creation support device for creating a high-quality web site in a short time by proposing a combination of parts optimized for each user.SOLUTION: A website creation assistance device 1 includes a web part management database 1212 in which web part information that is an element constituting a website is stored, a profile database 1211 in which profile information of a user 4 who creates the website is stored, and a communication unit 13 that enables communication of site-related information relating to the website with a user terminal 2 operated by the user 4. 1212, and a web site creation support part 11 for supporting the creation of the web site by presenting it to the user 4 as web creation support information.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a website creation support device. [Background technology]

[0002] Patent document 1 discloses a technique for a homepage design creation method that includes an answer receiving step for receiving answers to questions about products that the homepage to be created will handle, a combination extraction step for extracting, based on the answers, a combination of homepage parts corresponding to each answer pattern from a storage means that has registered combinations of homepage parts, and a design creation step for extracting, based on the combinations, parts from the storage means that has registered homepage parts, and combining the extracted parts to create the homepage design.

[0003] According to the technology of Patent Document 1, a combination of pre-prepared parts is selected based on the answers to questions about the attributes of the products to be handled on the homepage that the user is trying to create, and the homepage design is constructed based on that combination, making it possible to create a homepage cheaply and easily. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2002-351783 Summary of the Invention [Problem to be solved by the invention]

[0005] The technology in Patent Document 1 makes it possible to select a combination of pre-prepared parts in response to answers to questions and to build a homepage design based on that combination, but because it is limited to combinations of parts registered for each answer pattern to product-related questions, the combination of parts may not be optimal for creating the homepage that the user desires.If a combination of parts optimized for the user is not suggested, it becomes difficult to create a high-quality homepage in a short time, and improvements are desired.

[0006] The present invention has been made in consideration of these circumstances, and its purpose is to provide a website creation support device that enables the creation of high-quality websites in a short amount of time by proposing combinations of parts that are optimized for each individual user. [Means for solving the problem]

[0007] As a result of extensive research into solving the above-mentioned problems, the inventors discovered that the above-mentioned object can be achieved by predicting the website a user desires based on profile information and site-related information and presenting the user with appropriate web part information selection candidates. The inventors then completed the present invention. Specifically, the present invention provides the following:

[0008] The present invention includes a web part management database that stores information about web parts that are elements that make up a website; a profile database in which profile information of users who create the website is stored; a communication unit that enables communication of site-related information about the website with a user terminal operated by the user; The website creation support device has a website creation support unit that predicts the website desired by the user based on the profile information and the site-related information, extracts selection candidates for web part information suitable for creating the website from the web part management database, and presents them to the user as web creation support information to support the creation of the website.

[0009] With the above configuration, the system predicts the website desired by the user based on the profile information and site-related information, and then extracts appropriate web part information selection candidates from the web part management database and presents them to the user. This reduces the burden on users in selecting designs and functions, and provides suggestions optimized for each individual user, enabling the creation of high-quality websites in a short amount of time. [Effects of the Invention]

[0010] According to the present invention, a combination of parts optimized for each individual user is proposed, making it possible to create a high-quality website in a short time. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is an explanatory diagram showing the flow of information in a website creation support device according to this embodiment. [Figure 2] FIG. 2 is an explanatory diagram showing the contents of a data table of the web part management database of this embodiment. [Figure 3] FIG. 3 is an explanatory diagram showing the contents of the data table of the profile database of this embodiment. [Figure 4] FIG. 4 is an explanatory diagram showing the flow of information in the website creation support device of this embodiment. [Figure 5] FIG. 5 is an explanatory diagram showing the flow of information in the website creation support device of this embodiment. [Figure 6] FIG. 6 is an explanatory diagram showing the flow of information in the website creation support device of this embodiment. [Figure 7] FIG. 7 is an explanatory diagram showing the contents of the data table of the dropout rate database of this embodiment. [Figure 8] FIG. 8 is an explanatory diagram showing the flow of information in the website creation support device of this embodiment. [Figure 9] FIG. 9 is a flowchart of the website creation support program of this embodiment. [Figure 10] FIG. 10 is an explanatory diagram showing the process of creating a website according to this embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] First, although the following disclosure, diagrams, and / or claims may be described as being presented alone or in combination with one or more other aspects, the subject matter of the immediate disclosure is not intended to be so limited. That is, the immediate disclosure, diagrams, and claims are intended to encompass the various aspects described herein, each alone or in one or more combinations with each other. For example, even if the immediate disclosure describes and illustrates a first, second, and third embodiment in such a way that the first embodiment is described and illustrated specifically in conjunction with the second embodiment, or the second embodiment is described and illustrated only in conjunction with the third embodiment, the immediate disclosure and illustrations are not so limited and may include only the first embodiment, only the second embodiment, only the third embodiment, or one or more combinations of the first, second, and / or third embodiments, such as the first and second embodiments, the first and third embodiments, the second and third embodiments, or the first, second, and third embodiments.

[0013] The use of the phrase "or" in this document shall mean a "non-exclusive" arrangement unless expressly specified otherwise. For example, when we say "item x is A or B," we mean either: (1) item x is either A or B, but not both; or (2) item x is both A and B. In other words, the word "or" is not used to define an "exclusive" arrangement.

[0014] Additionally, the phrases "comprising at least one of" and "comprising at least one of the following," when used in conjunction with a system or element, mean that the system or element includes one or more of the elements listed after the phrase. For example, if there are three types of elements, element 1 through element 3, the phrases "comprising at least one of" and "comprising at least one of the following" are to be interpreted as any of the following structural arrangements: a device including the first element, a device including the second element, a device including the third element, a device including the first and second elements, a device including the first and third elements, a device including the second and third elements, or a device including the first, second, and third elements.

[0015] A similar interpretation is intended when the phrase "used in at least one of the following" is used in this context. Furthermore, as used in this context, "and / or" is used as a verbal conjunction to indicate that one or more of the listed elements or conditions are included or occur. For example, a device including a first element, a second element, and / or a third element is to be interpreted as any of the following structural arrangements: a device including the first element, a device including the second element, a device including the third element, a device including the first element and the second element, a device including the first element and the third element, a device including the second element and the third element, or a device including the first element, the second element, and the third element.

[0016] In addition, the use of the phrase "and / or" in the text means a "non-exclusive" agreement, as stipulated in the Japanese Industrial Standards (JIS) "Format and preparation method of standard sheets JIS Z 8301."

[0017] Hereinafter, an example of an embodiment of the present invention will be described in detail with reference to the drawings.

[0018] (Website Creation Support Device 1: Overview) As shown in Figure 1, the website creation support device 1 is configured to predict the website desired by the user 4 based on the profile information of the user 4 creating the website and site-related information about the website sent and received between the user 4 and the user terminal 2 operated by the user 4, and to support the creation of the website by presenting to the user 4 a selection of web part information suitable for creating the website as web creation support information.

[0019] Here, a "website" refers to a series of web pages published on the Internet 5. These web pages are created using web technologies such as HTML, CSS, and JavaScript and are accessible to users 4 through a browser. Websites are built for a variety of purposes, including providing information, services, entertainment, and e-commerce. A website typically consists of multiple pages, each connected to another through links. This allows users 4 to easily find information through navigation. The content of a web page can be provided in various formats, such as text, images, video, and audio. Websites are hosted on computer systems called web servers 3 and published on the Internet 5 through domain names. Users 4 can reach website pages by entering the domain name in the address bar of their browser or by accessing it through a search engine. Websites can also improve their rankings in search engine results for specific keywords or phrases through search engine optimization (SEO). This increases traffic to the website and promotes the use of targeted information and services.

[0020] "Web part information" refers to the elements that make up a website, i.e., information about the various components that make up a website. This information includes detailed data about the basic components that realize a website's design and functionality. Specifically, web part information includes a wide range of elements, such as text, images, buttons, navigation menus, forms, videos, icons, and banners. Each web part is assigned metadata such as part ID, part type, category, description, creation date, and update date, making it easy to identify and manage parts. Web part information serves to enhance the design and functionality of a website. For example, text parts include information such as font style, size, and color, ensuring design consistency. Image parts include information such as image size, resolution, and placement, allowing adjustments to enhance visual appeal. Interactive parts such as buttons and navigation menus include settings to improve the usability of the user interface. Furthermore, web part information can be customized according to User 4's needs. This allows for the creation of websites tailored to specific purposes and themes. Based on web part information, User 4 can freely combine web parts, adjust the layout, and create their own unique websites.

[0021] Web part information is information used to improve the functionality of a website. For example, form parts incorporate interactive features such as user input validation, data transmission destination settings, and user-friendly error message display. This allows for a comfortable website experience for website visitors. Video parts also include information for effectively displaying media content, such as video playback settings, thumbnail settings, and captions. This allows for easy integration of visually appealing content to attract visitors' attention. Web part information also contributes to search engine optimization (SEO). By providing appropriate metadata for each part, search engines can accurately understand the content of a website and rank it highly in search results. For example, providing appropriate alternative text for an image part can increase its visibility in image searches. Web part information also includes display settings for different devices (desktop, tablet, smartphone), allowing for optimal layouts for each device. It is also preferable for web part information to be integrated with a version control system to track the update history of parts and revert to previous versions as needed.

[0022] "User 4" refers to an individual or corporation that uses website creation support device 1 to build and manage a website. User 4 has various profile information depending on the purpose of creating a website, and can receive support for creating a website customized based on that information.

[0023] Specifically, user 4 first logs in to the website creation support device 1 and provides his or her profile information. This information includes his or her name, occupation, company name, and areas of interest. For example, if user 4 is a lawyer, he or she is required to enter detailed information such as an interest in business law and contracts. This profile information is recorded in the profile database 1211 and used by the website creation support unit 11. When user 4 begins building a website, the website creation support unit 11 extracts optimal web part information from the web part management database 1212 based on user 4's profile information and site-related information. This allows user 4 to select web parts that best suit his or her needs. For example, if user 4 is a lawyer providing legal services, a catchphrase and profile parts emphasizing his or her reliability and expertise are suggested.

[0024] Furthermore, user 4 can easily operate the system through a user interface (UI) for adding specific designs and content to each page of the website. Websites can be constructed intuitively using drag-and-drop operations and text-generating input forms. This makes it possible to create high-quality websites without any coding knowledge. When user 4 publishes the website, he or she can utilize the deployment and update functions to quickly and efficiently deploy the site on the Internet 5. Furthermore, if the website needs updating or maintenance even after publication, user 4 can access the web part management database 1212 to add new parts or edit existing parts.

[0025] "Profile Information" refers to detailed information about an individual or corporation that User 4 provides to build and manage a website. Profile Information includes User 4's name and occupation as basic identifying information. For example, if User 4 is a lawyer named "Yamada Taro," his name and occupation are recorded as part of his profile information. It also includes the name of the company or organization to which User 4 belongs, which clarifies the background of User 4's business or services. For example, if User 4 belongs to "Yamada Law Office," this information is recorded.

[0026] Furthermore, the profile information includes the user's 4 areas of interest. This allows the website creation support device 1 to propose web parts and design templates tailored to the user's 4's specific needs and preferences. For example, for a user 4 interested in business law and contracts, related legal content and contract templates are preferentially proposed. The profile information is recorded in the profile database 1211 of the website creation support device 1. The profile database 1211 is designed to centrally manage detailed information for each user 4 and enable quick access as needed. This allows the user 4 to receive optimal support based on their individual profile information when building a website. For example, the profile information plays an important role in selecting and customizing web parts. Appropriate catchphrases, service descriptions, profile parts, etc. are proposed based on the user 4's occupation and areas of interest. Parts that emphasize the user 4's expertise and experience are also provided. This allows the user 4 to build a website that effectively highlights their strengths and characteristics.

[0027] "Site-related information" refers to various data necessary for creating and managing a website. This information includes detailed data about the components, design, and functions of a website. Specifically, site-related information includes information about the basic structure of a website. For example, detailed design information for each page of a website, such as the layout of each web page, the link structure between pages, the placement of navigation menus, service introduction pages, and inquiry pages.

[0028] As a concrete example, in the case of the Yamada Law Office website, this includes information such as the site name, site purpose, theme, color scheme, font, main menu items, homepage content, service content page, lawyer introduction page, testimonials page, contact page, blog, FAQ, and social media integration. This ensures a consistent structure throughout the site, providing visitors with easy-to-understand navigation.

[0029] Site-related information also includes data about content. Specifically, it includes all content elements displayed on a web page, such as text, images, videos, buttons, and forms. Each content element includes detailed information such as its specific content, placement, and style settings. For example, the image slider, tagline, service overview, and customer testimonials displayed on the top page are saved as site-related information.

[0030] Furthermore, site-related information also includes information about the website design, such as color settings, font styles, layout patterns, and theme settings. This helps maintain consistency in the overall design of the website and strengthen the brand image for users 4. For example, Yamada Law Office uses a color scheme of navy blue, white, and gray, and sans-serif fonts (e.g., Arial, Helvetica), to achieve a professional, reliable, and simple design.

[0031] Site-related information also includes data related to SEO (Search Engine Optimization). For example, meta tags, keywords, page titles, descriptions, etc. are stored to effectively communicate the content of a website to search engines. This helps improve a website's search ranking and attract more visitors.

[0032] Site-related information is also used for the operation and maintenance of the website. For example, it includes information for collecting access analysis data and visitor behavior data and evaluating website performance. This makes it possible to analyze visitor behavior patterns and identify areas for improvement on the website. Site-related information is stored in the site-related database 1213. The site-related database 1213 is designed to centrally manage all information necessary for creating and managing the website and to allow users 4 to quickly access the information they need. This allows the website to be built and operated efficiently and effectively.

[0033] (Website creation support device 1: memory unit 12) The website creation support device 1 has a memory unit 12. The memory unit 12 has a function for storing various data that constitute a website and enabling quick access when needed. The main function of the memory unit 12 is to safely and efficiently store information including a web part management database 1212, a profile database 1211, and various other databases, and to provide this data so that it can be accessed by the website creation support unit 11, etc. This makes it possible to provide the information required by the user 4 quickly and accurately.

[0034] More specifically, the memory unit 12 has a storage device for saving data. The storage device may be a solid-state drive (SSD) or a hard disk drive (HDD). These devices have the ability to quickly read and write large amounts of data, making it possible to efficiently manage large amounts of data such as web part information and profile information.

[0035] The storage unit 12 also has a RAM (random access memory) for temporarily storing data that requires fast access. The RAM temporarily stores data that the processor accesses frequently, improving the speed of data processing. The storage unit 12 also has a database management system (DBMS). The database management system (DBMS) reads and writes data using SQL (Structured Query Language) and plays a role in maintaining data integrity and consistency. This ensures that information stored in databases such as the web part management database 1212 and the profile database 1211 is kept accurate and up-to-date.

[0036] Furthermore, the storage unit 12 is equipped with a backup system that periodically backs up data in case of data loss or corruption. The backup system saves a copy of the data stored in the storage device in another location, enabling data recovery. This ensures data integrity even in the unlikely event of a failure. It is preferable that the storage unit 12 is equipped with security functions to prevent unauthorized access or leakage of data. Specifically, these functions include data encryption, access control, and audit log recording. This protects the confidential information and important data of the user 4, achieving highly reliable data management.

[0037] (Website creation support device 1: storage unit 12: web parts management database 1212) The storage unit 12 configured as described above has a web part management database 1212 that stores information about web parts, which are elements that make up a website. The web part management database 1212 is a database for centrally managing information about various web parts that make up a website. This database 1212 stores detailed information and metadata about the web parts. The web part management database 1212 includes a "customer information table," a "web part information table," and a "part usage status table." The customer information table records information about the user 4 who creates the website, including the customer ID, name, email address, account creation date, and part of profile information. This information enables customization and personalization for each user 4.

[0038] The web part information table stores detailed information about each web part. Specifically, it includes metadata such as part ID, part type, content, version, creator, creation date, and tags. This clarifies the characteristics and uses of each web part, allowing for the quick selection of the appropriate web part as needed. The part usage table tracks how each web part is used by which user 4. It contains information such as customer ID, part ID, usage location ID, number of uses, and last use date, allowing for the evaluation of the frequency and performance of web parts. By combining these tables, the web part management database 1212 improves the efficiency and accuracy of the website creation process, increases the reusability of web parts, and supports the creation of websites without waste. It also facilitates the suggestion of web parts based on user 4 profiles and the management of information needed for site updates and maintenance.

[0039] To explain the web part management database 1212 in more detail using a specific example, as shown in Figure 2, the customer information table records information about users 4 who create websites. For example, Yamada Taro (customer ID 1) is a lawyer and belongs to Yamada Law Office. Tanaka Hanako (customer ID 2) is a designer and belongs to Tanaka Design Studio. The interests and concerns of each user are also recorded, making it possible to customize and personalize the website based on the preferences and needs of user 4.

[0040] The Web Part Information Table stores detailed information about each Web Part. For example, Part ID WP001 is a catchphrase, "Yamada Law Office, trusted and proven," created by Yamada Taro. Part ID WP002 is a service description, "Legal consultation: We solve your legal problems," also created by Yamada Taro. This information clarifies the characteristics and uses of each Web Part, allowing you to quickly select the appropriate Web Part as needed.

[0041] The part usage status table is used to track how each web part was used and by which user 4. For example, part ID WP001 used by Yamada Taro (customer ID 1) was used three times on PAGE001, and the last usage date was January 15, 2023. Also, part ID WP004 used by Tanaka Hanako (customer ID 2) was used once on PAGE004, and the last usage date was January 22, 2023. This makes it possible to evaluate the usage frequency and performance of parts.

[0042] (Website creation support device 1: memory unit 12: profile database 1211) As shown in FIG. 1 , the storage unit 12 has a profile database 1211 that stores profile information about users 4 who create websites. The profile database 1211 is a database for centrally managing detailed profile information about users 4 who create websites. The profile database 1211 stores various data necessary for website creation, such as personal information, occupational information, and areas of interest of the users 4, enabling the provision of website creation support customized for each user 4. The profile database 1211 contains basic identification information about the users 4. Specifically, the user 4's name, occupation, company name, contact information, and the like are recorded. For example, for a lawyer named Yamada Taro, the profile database 1211 stores information about Yamada Taro's name, occupation as a lawyer, and the Yamada Law Firm to which he belongs.

[0043] Furthermore, the profile database 1211 also contains information about the user 4's areas of interest. This information includes the user 4's fields of expertise, hobbies, and topics of interest. For example, the profile of a user 4 who is interested in business law and contracts will include information about business law and contracts. This information plays an important role when the website creation support unit 11 suggests optimal web parts and design templates for the user 4. The profile database 1211 functions in conjunction with each unit of the website creation support device 1. For example, profile information received from the user 4's terminal via the communication unit 13 is recorded in the profile database 1211. The information in the profile database 1211 is also used when extracting appropriate web part information from the web part management database 1212. This enables customized support for each user 4, enabling websites to be created according to individual needs.

[0044] To explain this using a specific example, as shown in Figure 3, the profile database 1211 includes customer ID, name, age, gender, occupation, company name, service content, interests, and past usage history. For example, Yamada Taro, customer ID 1, is a 35-year-old man who works as a lawyer at Yamada Law Office. His service content is legal consultation, and he is interested in business law and contracts. He has also used the website creation support device three times in the past. Tanaka Hanako, customer ID 2, is a 28-year-old woman who works as a designer at Tanaka Design Studio, where she provides graphic design. She is interested in UI / UX design and branding, and has used the support device once in the past.

[0045] The profile database 1211 includes information on user 4's areas of interest in addition to basic identification information. That is, the user 4's areas of expertise, hobbies, topics of interest, etc. are recorded. For example, Suzuki Ichiro (customer ID 3) is a 45-year-old man who works as an engineer at Suzuki Engineering and is in charge of system development. He is interested in AI and IoT, and has used the website creation support device 1 five times in the past.

[0046] (Website creation support device 1: User terminal 2) As shown in FIG. 1, a website creation support device 1 is capable of communicating data with a user terminal 2. The user terminal 2 is a device such as a computer, smartphone, or tablet that a user 4 uses to create and manage a website. The user terminal 2 functions as an interface for communicating with the website creation support device 1, providing an environment in which the user 4 can intuitively manipulate the website design and content. The user terminal 2 establishes a connection with the website creation support device 1. This connection is made via the Internet 5, and can be accessed through a web browser or dedicated application on the user terminal 2. This allows the user 4 to use various functions of the website creation support device 1.

[0047] The user terminal 2 has a user interface (UI) function. The UI is a screen that allows the user 4 to operate the design and content of a website, and is designed to be intuitively operated through drag-and-drop operations and input forms. For example, the user 4 can use a text generation input form to enter text to be posted on a web page and an image input form to upload images. The user terminal 2 also provides tools for selecting and customizing web parts. Based on web part information from the web part management database 1212, the user 4 can select parts that suit their preferences and needs and add them to the page. This allows the user 4 to efficiently build a website that is optimal for their business or service.

[0048] The user terminal 2 also provides a real-time preview function, allowing the user 4 to instantly check the results of any changes made to the website. For example, if the user adjusts the placement or style of a web part, the user 4 can see in real time how the changes will be displayed. This makes it possible to improve the consistency and quality of the design. The user terminal 2 also has authentication and authorization functions for the user 4, which prevent unauthorized access and ensure the safety of the user 4's data. Communication between the user terminal 2 and the website creation support device 1 is encrypted, ensuring thorough data protection.

[0049] (Website creation support device 1: communication unit 13) The website creation support device 1 has a communication unit 13 that enables communication of site-related information about a website with a user terminal 2 operated by a user 4. The communication unit 13 has a communication function that establishes communication with the user terminal 2 operated by the user 4. The communication unit 13 is capable of communicating with the user terminal 2 via a data communication network such as the Internet 5. Here, examples of the "data communication network" include the Internet 5, an intranet, an extranet, a mobile communication network, a wide area network (WAN), a local area network (LAN), a metropolitan area network (MAN), and a satellite communication network.

[0050] The communication unit 13 is constructed with a highly reliable hardware configuration to reliably establish communication with the user terminal 2 and ensure smooth information exchange. Specifically, it is equipped with a network interface card (NIC) that ensures high-speed data transfer and stable connection, and minimizes the risk of delays and data loss. The communication unit 13 also includes routers and switches that efficiently route and distribute data packets and optimize network traffic management. Furthermore, the communication unit 13 is equipped with a firewall and intrusion detection system (IDS) that protect the network from external unauthorized access and internal security threats, enhancing data protection for the user 4.

[0051] Note that data communication between the communication unit 13 and the user terminal 2 is not limited to a data communication network such as the Internet 5, and Bluetooth (registered trademark), a wireless communication standard for short-range communication, may also be used. Furthermore, data communication may be via a dedicated line to prevent information leakage. Furthermore, when data communication is performed between the communication unit 13 and the user terminal 2 via the Internet 5, it is preferable that the communication unit 13 and the user terminal 2 have a VPN (Virtual Private Network) function. If a VPN function is incorporated into the communication unit 13 and the user terminal 2, for example, data communication from the user terminal 2 passes through a VPN connection, thereby protecting the data from unauthorized external access and ensuring secure communication.

[0052] The communication unit 13 configured as described above allows the user 4 to manipulate the design and content of the website in real time. By connecting the communication unit 13 to the user terminal 2, the user 4 can send various website information and receive necessary data and support information. For example, data entered by the user 4 to build a website and information about selected web parts are sent to the website creation support unit 11 via the communication unit 13 and stored in the web part management database 1212. Similarly, web part selection candidates and design suggestions generated by the website creation support unit 11 are also sent to the user terminal 2 via the communication unit 13.

[0053] The communication unit 13 also has a function for real-time data synchronization, which allows the user 4 to instantly reflect changes to the website design and content. For example, if the user 4 makes changes to the layout, the changes are updated in real time via the communication unit 13 and displayed on the user terminal 2. This allows the user 4 to immediately check the results of the changes and quickly make any necessary corrections.

[0054] (Website Creation Support Device 1: Website Creation Support Unit 11) Website creation support device 1 has a control unit (computer) that is a website creation support unit 11. Website creation support unit 11 has the function of predicting the website desired by user 4 based on profile information and site-related information, extracting selection candidates for web part information suitable for website creation from web part management database 1212, and presenting this to user 4 as web creation support information, thereby supporting website creation.

[0055] Here, "guessing the website desired by user 4" is performed through multiple steps. Specifically, user 4's profile information is collected. This information includes name, occupation, company name, and areas of interest. For example, if user 4 is a lawyer and is interested in business law, this information is stored in profile database 1211. Next, site-related information is collected. For example, this includes details such as the purpose of the website user 4 wants to create, target audience, design preferences, and required functions. After this information is collected, the profile information and site-related information are analyzed to guess the website desired by user 4. This process uses large-scale language models (LLMs) and machine learning algorithms.

[0056] "Web part information selection candidates suitable for website creation" is a list of suggested web parts that are most suitable for the characteristics and purpose of the website desired by User 4. These selection candidates are generated based on User 4's profile information and site-related information, and are elements for optimizing the design and functionality of the website. For example, if User 4, a lawyer, desires a website focused on business law, the suggested selection candidates include a catchy slogan that emphasizes reliability, a description of legal consultation services, and profile parts that highlight expertise. Parts that display customer feedback and ratings are also suggested selection candidates.

[0057] To explain the website creation support unit 11 in more detail, the website creation support unit 11 has a support function for enabling the user 4 to effectively and efficiently build and manage a website. The website creation support unit 11 selects optimal web parts and proposes website designs based on the user 4's profile information and site-related information. Specifically, the website creation support unit 11 collects information such as the user 4's name, occupation, company name, and areas of interest as profile information, and collects detailed site-related information about the website the user 4 wants to create. Based on the collected profile information and site-related information, the website creation support unit 11 predicts the website the user 4 desires and generates a selection of optimal web part information. This generation process uses a web part selection algorithm using a large-scale language model (LLM) or a machine learning model. The web part selection algorithm will be described in detail later. This allows the website creation support unit 11 to deeply understand the user 4's intentions and preferences and propose web parts and templates that provide optimal designs and functions.

[0058] For example, if user 4 is creating a website for a law firm, the system suggests options such as a catchy slogan that emphasizes reliability, profile parts that highlight expertise, and parts that display customer feedback. Effective combinations of parts are also suggested based on user 4's past selections and success stories of other similar users. Website creation support unit 11 transmits information about the selected web parts as web creation support information to user terminal 2 via communication unit 13. User 4 can then proceed with building their website based on this web creation support information. In other words, user 4 can easily arrange and customize the suggested parts using drag-and-drop operations or input forms.

[0059] Furthermore, the website creation support unit 11 provides a real-time design preview function, which allows changes made by the user 4 to be reflected immediately, allowing the user 4 to proceed with the creation process while checking the appearance and functions of the website. This allows the user 4 to complete the website in a form that is closest to their vision.

[0060] (Website creation support device 1: Website creation support unit 11: Web part selection algorithm) Examples of web part selection algorithms include rule-based algorithms, machine learning algorithms based on machine learning models, recommendation algorithms, attribute-based selection algorithms, large-scale language algorithms based on large-scale language models, etc. By combining these algorithms, it becomes possible to select optimal web parts that meet the diverse needs of users 4.

[0061] In more detail, the rule-based algorithm analyzes User 4's profile information and site-related information based on a set of predefined rules and conditions to predict the characteristics of the website User 4 desires. For example, it extracts website characteristics related to a specific profession or area of ​​interest and generates optimal web part information selection candidates based on those characteristics. This method is characterized by its simplicity, ease of understanding, and ability to reliably respond to specific conditions. For example, if User 4 is a "lawyer," it predicts a website structure that emphasizes "trustworthiness" and "professionalism," and selects related catchphrases and legal advice parts.

[0062] The machine learning algorithm learns from past data, finding patterns and trends to make predictions. The algorithm analyzes User 4's profile information and site-related information to infer the characteristics of the website User 4 desires. For example, it generates optimal web part information selection candidates based on data on websites User 4 has created in the past and other similar users. This allows it to prioritize suggestions of web parts frequently used by User 4 and web parts highly rated by other similar users.

[0063] The recommendation algorithm predicts the characteristics of the website desired by User 4 based on User 4's preferences and behavioral history. The algorithm learns User 4's tendency to prefer specific designs and content, and generates web part information selection candidates based on that. For example, it can refer to designs that User 4 has used favorably in the past, or parts that other Users 4 have selected under similar conditions.

[0064] The attribute-based selection algorithm infers the characteristics of the website desired by User 4 based on the attribute information of each web part (occupation, interests, age, etc.). For example, it generates optimal web part information selection candidates by taking into account attributes such as the design, function, frequency of use, and ratings of the web part. This makes it possible to provide the web part that is most suitable for User 4's specific requests and conditions.

[0065] The large-scale language algorithm uses a large-scale language model to analyze user 4's profile information and site-related information and infer the characteristics of the website desired by user 4. For example, if user 4 inputs, "I want to create a website for a reliable law firm," the large-scale language model extracts keywords such as "reliability" and "law firm" and infers the structure of the related website. Based on this, it generates selection candidates for optimal web part information. Details of website creation support unit 11 that applies the large-scale language algorithm will be described later.

[0066] (Website Creation Support Device 1: Actions and Effects) As described above, website creation support device 1 has web part management database 1212 storing web part information, which is the element that makes up a website; profile database 1211 storing profile information of user 4 who creates the website; communication unit 13 that enables communication of site-related information regarding the website between user terminal 2 operated by user 4; and website creation support unit 11 that predicts the website desired by user 4 based on the profile information and site-related information, extracts from web part management database 1212 selection candidates for web part information suitable for creating the website, and presents this to user 4 as web creation support information to support the creation of the website.

[0067] In the above-described website creation support device 1, the web part management database 1212 stores information on various web parts that make up a website, thereby providing a wide range of materials for realizing diverse designs and functions. Furthermore, the profile database 1211 records profile information about the user 4, such as the age, gender, occupation, and interests, making it possible to accurately grasp the preferences and needs of the user 4. The communication unit 13 has a function that enables bidirectional communication of site-related information with the user terminal 2 operated by the user 4, enabling real-time information exchange. This allows the user 4 to receive feedback at any time during the website creation process, enabling quick and flexible response.

[0068] The website creation support unit 11 has the function of predicting the website desired by the user 4 based on the profile information and site-related information, and extracting appropriate web part information selection candidates from the web part management database 1212. This allows the user 4 to easily build a website that meets his or her needs. The website creation support unit 11 accurately reflects the preferences and needs of the user 4 and proposes the optimal combination of web parts, thereby realizing efficient and accurate website creation. With this configuration, the website creation support device 1 can automatically select the optimal web parts based on the user 4's profile information and site-related information and present them to the user 4. As a result, the burden on the user 4 in selecting designs and functions is reduced, and proposals optimized for each individual user 4 are made, allowing a high-quality website to be created in a short amount of time.

[0069] (Website Creation Support Device 1: Website Creation Support Unit 11: Example of Application of Large-Scale Language Model) The website creation support device 1 configured as above may be configured such that the website creation support unit 11 includes a large-scale language model, and the large-scale language model is used to propose web part selection candidates.

[0070] Specifically, as shown in FIG. 4, the website creation support unit 11 has a large-scale language model unit 1131 having a large-scale language model that combines one or more types of natural language processing to probabilistically predict how likely words and sentences given in a prompt are to occur in natural language, and a selection candidate creation prompt generation unit 1132 that, when site-related information is received from the user terminal 2 via the communication unit 13, includes the site-related information in the prompt of the large-scale language model unit 1131, predicts the website desired by the user 4 based on the web part information in the web part management database 1212 and the profile information in the profile database 1211, and causes the large-scale language model to generate selection candidates for web part information suitable for creating a website as web creation support information.

[0071] (Website creation support device 1: Website creation support unit 11: Large-scale language model unit 1131) The large-scale language model unit 1131 includes a large-scale language model. Here, a "large-scale language model" is a type of probabilistic model used in natural language processing, which is a model for probabilistically predicting how likely a given word or sentence is to occur in natural language. Specifically, the language model calculates the occurrence probability of a given word string or sentence, or compares the occurrence probabilities of multiple word strings or sentences, thereby enabling the automatic generation of the most likely word or sentence based on the context when predicting the next word or sentence, or the generation of a sentence that satisfies specific conditions. That is, the large-scale language model is a language model that combines one or more natural language processing processes, such as morphological analysis, syntactic analysis, semantic analysis, context analysis, and intention analysis, to probabilistically predict how likely a word or sentence given in a prompt is to occur in natural language. The large-scale language model unit 1131 is configured to analyze a prompt, the content of which instructs the generation of web creation support information, such as web part information selection candidates, using the large-scale language model, and to predict and generate web creation support information based on the content of the analyzed prompt.

[0072] "Natural language processing" enables a computer to understand text and audio data written in natural language and execute processing appropriate to the purpose. Specific examples include morphological analysis, which breaks natural language down into "morphemes," the smallest units that make up the language, and assigns information such as parts of speech; syntactic analysis, which analyzes the grammatical structure of natural language to clarify the structure and meaning of a sentence; semantic analysis, which analyzes the meaning of natural language to understand the meaning of words and sentences and make logical judgments and inferences; contextual analysis, which understands natural language while taking into account the context before and after a sentence; and intent analysis, which extracts the intention of a speaker or writer from a conversation or text using natural language. "Natural language processing" processes natural language by combining processes such as morphological analysis, syntactic analysis, semantic analysis, contextual analysis, and intent analysis, and enables the generation of web development support information, machine translation, automatic summarization, question-answering systems, and speech recognition that support collaborative activities in this embodiment.

[0073] A "prompt" is a word or sentence input to the large-scale language model unit 1131, and serves as the starting point for the large-scale language model to generate web creation support information. The generation of web creation support information using a large-scale language model based on a prompt is a major difference from regular (conventional) machine learning. In regular machine learning, a model learns from training data and predicts output data for input data. For example, a machine learning model for image recognition learns from training data of images of cats and dogs and classifies the input image as either a cat or a dog. On the other hand, when generating web creation support information using a large-scale language model, the model learns not only from training data but also from prompts that provide instructions and information regarding the output data that the model is desired to generate.

[0074] In more detail, while conventional machine learning models can only generate content contained in the training data, large-scale language models can use prompts to generate new content not contained in the training data. For example, even if the training data only contains content about IT project managers, prompts can also generate content about marketing project managers. Furthermore, while conventional machine learning models can only generate variations of content contained in the training data, large-scale language models can use prompts to generate new variations of output data not contained in the training data. For example, even if the training data only contains content about technical skills, prompts can also generate content about soft skills. Furthermore, while conventional machine learning models can only generate new content by recursively combining content contained in the training data, large-scale language models can use prompts to generate creative content not contained in the training data. For example, even if the training data only contains content about the relationship between an IT engineer and User 4, prompts can generate completely new perspectives, such as new ideas based on the relationship between the IT engineer and User 4.

[0075] Furthermore, it is preferable that the large-scale language model in the large-scale language model unit 1131 is specialized for creating web creation support information. That is, it is preferable that the large-scale language model is connected to a website database that stores data related to websites and is specialized for creating web creation support information by constantly learning. The website database stores a large amount of data on areas important to website creation, such as web design, user experience (UX), content creation, and SEO (search engine optimization). Specifically, it includes data such as success stories, best practices, the latest design patterns, and user feedback. By utilizing this data, the large-scale language model can make more accurate and appropriate suggestions.

[0076] (Website creation support device 1: Website creation support unit 11: Selection candidate creation prompt generation unit 1132) When the selection candidate creation prompt generation unit 1132 receives site-related information from the user terminal 2 via the communication unit 13, it includes the site-related information in the prompt of the large-scale language model unit 1131, predicts the website desired by the user 4 based on the web part information in the web part management database 1212 and the profile information in the profile database 1211, and has the function of generating selection candidates for web part information suitable for creating a website in the large-scale language model as web creation support information.

[0077] The selection candidate creation prompt generation unit 1132 exists as a module independent of the large-scale language model unit 1131. The reason for this is that the selection candidate creation prompt generation unit 1132 has the role of generating appropriate prompts based on site-related information, which is different from the natural language processing role of the large-scale language model unit 1131. For this reason, treating the two as independent modules makes it possible to clearly distinguish between their respective roles and facilitate system design and maintenance. Furthermore, implementing the selection candidate creation prompt generation unit 1132 as an independent module makes it possible to flexibly respond when the large-scale language model is updated.

[0078] The selection candidate creation prompt generation unit 1132 has an interface for accepting site-related information received from the user terminal 2 and prompt generation logic for generating a prompt with content instructing the creation of web creation support information based on the received site-related information. The prompt generation logic is used to select a prompt template appropriate for the web creation support information. Conditional branching is used for this selection, and a prompt template that matches the web creation support information is selected. The selection candidate creation prompt generation unit 1132 also has a prompt transmission interface for transmitting the generated prompt to the large-scale language model unit 1131.

[0079] The prompt templates for the selection candidate creation prompt generation unit 1132 are prepared in advance. The prompt templates provide the document structure and content framework that form the basis for the selection candidate creation prompt generation unit 1132 to effectively create web creation support information. These templates are customized and designed in advance according to the needs and circumstances of the user 4. Specifically, the prompt templates predefine questions and information formats that match the interests and intentions of the user 4. For example, if the user 4 is a lawyer and is creating a website for his or her law firm, the prompt template would include items such as a "catchphrase that emphasizes reliability," "detailed explanation of legal consultation services," and "introduction of success cases." By using this prompt template, the specific web part information required by the user 4 can be effectively extracted.

[0080] Furthermore, the selection candidate creation prompt generation unit 1132 combines User 4's profile information and site-related information to grasp the overall picture of the website intended by User 4 and generate prompts that match that intent. For example, if User 4 desires a "casual legal consultation website targeted at young people," the prompt template is customized to include features such as "friendly design elements," "social networking features," and "enhanced FAQ section." The prompt generation logic of the selection candidate creation prompt generation unit 1132 selects the prompt template that best suits User 4's needs through complex conditional branching. In this process, more accurate prompts can be generated based on information learned from past data and similar cases. By sending this prompt to the large-scale language model unit 1131, specific web part information required for website creation is generated.

[0081] The prompt sending interface has the function of quickly and accurately sending the generated prompts to the large-scale language model unit 1131. The prompt sending interface allows prompts to be sent and received smoothly, enabling real-time information processing. This allows user 4 to efficiently create a website. As described above, the selection candidate creation prompt generation unit 1132 generates appropriate prompts based on the needs of user 4 and works in cooperation with the large-scale language model unit 1131 to provide web creation support information, thereby effectively supporting user 4 in creating a website.

[0082] The selection candidate creation prompt generator 1132 may be constructed using a rule-based logic circuit or a machine learning model. In the rule-based case, the prompt is generated using explicit conditional branching and transition rules. On the other hand, in the case of using a machine learning model, more flexible prompt generation becomes possible using a model learned from data (information) in the web part management database 1212.

[0083] Preferably, the selection candidate creation prompt generator 1132 is capable of receiving various information stored in the profile database 1211 and the web part management database 1212, and based on this information, the needs of user 4 are continuously evaluated and the prompt template is updated accordingly. In this case, for example, if user 4 adds new interests or concerns to his or her profile information, the prompt template is updated accordingly. This update reflects the information entered by user 4 in real time, enabling the selection candidate creation prompt generator 1132 to generate prompts based on the latest information. For example, if user 4 recently expressed interest in "online seminars" or "webinars," the corresponding prompt template is automatically updated to include items such as "online seminar announcements," "webinar event information," and "participant reviews."

[0084] In this way, the selection candidate creation prompt generation unit 1132 can continuously learn user 4's profile information and past site-related information and flexibly update the prompt template based on the results. This more accurately reflects the information provided by user 4 when creating a website, making it possible to select web part information that meets individual needs. Furthermore, the selection candidate creation prompt generation unit 1132 can optimize the prompt template based on the usage history and feedback of web parts selected by user 4 in the past. For example, if a specific web part receives high ratings, the template related to that part is enhanced and recommended to other users 4. Furthermore, for parts that receive low ratings, the cause is analyzed and the template content is modified as necessary.

[0085] In this way, the selection candidate creation prompt generation unit 1132 can improve the quality of website creation support information by continually improving the accuracy and applicability of the prompt templates, reflecting the trends and feedback of the user 4. This enables the user 4 to efficiently select web part information that best suits their needs and quickly create a high-quality website.

[0086] According to the above configuration, as shown in FIG. 5, in the website creation support device 1, the large-scale language model unit 1131 and the selection candidate creation prompt generation unit 1132 learn using information from the memory unit 12 equipped with each of the databases 1211 to 1216, and the site-related information input through the communication unit 13 corresponds to an explanatory variable in machine learning, and in response to a prompt from the selection candidate creation prompt generation unit 1132, the large-scale language model unit 1131 generates web creation support information as an objective variable in machine learning.

[0087] As a result, website creation support device 1 can integrate information from web part management database 1212 and profile database 1211 by including the site-related information entered by user 4 in the prompts of the large-scale language model, and generate web part selection candidates that are optimal for user 4. The web part selection candidates are sent to user terminal 2 via communication unit 13 as web creation support information, allowing user 4 to check and select the suggested web parts in real time. This enables user 4 to create a website intuitively and efficiently.

[0088] In this way, when using a large-scale language model, the website creation support device 1 provides advanced functions that are difficult to achieve with conventional machine learning algorithms. Specifically, it utilizes the unique capabilities of large-scale language models, such as complex text generation using natural language processing and context-based content generation. Conventional machine learning models typically perform prediction, classification, clustering, and other tasks using primarily numerical and categorical data, and are limited in their ability to process large amounts of text data and generate new text. On the other hand, large-scale language models are adept at learning language patterns from large amounts of text data and generating new text based on given prompts. Therefore, in situations requiring the use of diverse and complex language, such as web creation support information, an approach using a large-scale language model is more appropriate, and it is possible to provide functions that are difficult to achieve with conventional machine learning methods alone.

[0089] (Website creation support device 1: Website creation support unit 11: Proposal information creation prompt generation unit 1133) As shown in FIG. 4, it is preferable that the website creation support unit 11 further includes a suggested information creation prompt generation unit 1133 that, when selection candidates for web part information are generated, includes site-related information in the prompt of the large-scale language model unit 1131, and causes the large-scale language model to generate suggested information as web creation support information indicating a combination of selection candidates for web part information that will result in a website that user 4 prefers, based on the profile information in the profile database 1211.

[0090] After the selection candidates for web part information have been generated, the proposal information creation prompt generation unit 1133 has a function of including site-related information for the user 4 in the prompt of the large-scale language model unit 1131 to generate more detailed proposal information. Unlike the selection candidate creation prompt generation unit 1132, this proposal information creation prompt generation unit 1133 has a role of generating prompts for making suggestions regarding the specific structure and design of a website. In particular, it has a function of causing the large-scale language model to generate proposal information indicating how to combine the selection candidates for web part information to create a more effective website.

[0091] More specifically, the suggested information creation prompt generation unit 1133 grasps the overall picture of the websites preferred by User 4 based on User 4's profile information and site-related information, and generates prompts that match that user's intentions. For example, if User 4 desires a "casual legal consultation site targeted at young people," the prompt generation unit generates suggested information including such content as "friendly design elements," "SNS linkage functionality," and "enhanced FAQ section." This suggested information indicates how to combine candidate web part information options to create a more effective website.

[0092] Specifically, the suggested information creation prompt generation unit 1133 acquires user 4's profile information (e.g., occupation, interests, past site usage history, etc.) and site-related information (e.g., site purpose, target audience, required functions, etc.) in order to grasp the overall picture of the website desired by user 4. Based on this, the suggested information creation prompt generation unit 1133 extracts selection candidates for web part information that are most suitable for user 4's needs and creates a prompt for analyzing how to combine them effectively.

[0093] When the large-scale language model is given prompts on how to combine the web part information options, it generates suggestions such as: "On the homepage, place a catchy header incorporating 'friendly design elements' and a concise introduction that immediately draws users in. Next, place a 'detailed description of legal consultation services' in the main content area, highlighting specific cases and achievements. Also, display 'introductions to success stories' in the sidebar to make visitors feel trustworthy. Furthermore, incorporate an 'enhanced FAQ section' at the bottom of the page to allow visitors to quickly obtain information. Finally, include 'social media integration features' on all pages to allow visitors to easily share."

[0094] The proposal information creation prompt generation unit 1133 exists as a module independent of the large-scale language model unit 1131. The reason for this is that the proposal information creation prompt generation unit 1133 has a role of generating appropriate prompts based on site-related information, which is different from the natural language processing role of the large-scale language model unit 1131. For this reason, treating the two as independent modules makes it possible to clearly distinguish their respective roles and facilitate system design and maintenance. In addition, implementing the proposal information creation prompt generation unit 1133 as an independent module makes it possible to flexibly respond when the large-scale language model is updated.

[0095] The proposal information creation prompt generation unit 1133 has an interface for accepting site-related information received from the user terminal 2 and prompt generation logic for generating a prompt with content instructing the creation of web creation support information based on the received site-related information. The prompt generation logic is used to select a prompt template appropriate for the web creation support information. Conditional branching is used for this selection, and a prompt template that matches the web creation support information is selected. The proposal information creation prompt generation unit 1133 also has a prompt transmission interface for transmitting the generated prompt to the large-scale language model unit 1131.

[0096] The prompt templates for the proposal information creation prompt generation unit 1133 are prepared in advance. The prompt templates provide the proposal information creation prompt generation unit 1133 with a framework for document structure and content that serves as the basis for effectively creating web creation support information. These templates are customized and designed in advance according to the needs and circumstances of the user 4. The prompt sending interface also has the function of quickly and accurately sending the generated prompts to the large-scale language model unit 1131. The prompt sending interface allows prompts to be sent and received smoothly, enabling information processing in real time. This allows the user 4 to efficiently proceed with creating their website.

[0097] As described above, the suggested information creation prompt generation unit 1133 generates appropriate prompts based on the needs of the user 4 and cooperates with the large-scale language model unit 1131 to provide web creation support information, thereby enabling effective support for the creation of a website by the user 4. Furthermore, by having the large-scale language model generate suggested information indicating how to combine web part information selection candidates to create a more effective website, the creation of a high-quality website that meets the expectations of the user 4 is supported. In other words, by including the suggested information creation prompt generation unit 1133, the website creation support device 1 gains the added function of suggesting not only web part selection candidates but also their combinations, thereby enabling the provision of more accurate design proposals to the user 4. This significantly improves the convenience for the user 4 and further improves the efficiency of website creation.

[0098] (Website creation support device 1: Website creation support unit 11: Text information creation prompt generation unit 1134) It is preferable that the website creation support unit 11 has a text information creation prompt generation unit 1134 that, when selection candidates for web part information are generated, includes site-related information in the prompt of the large-scale language model unit 1131 and causes the large-scale language model to generate text information suitable for the website as web creation support information based on the profile information in the profile database 1211.

[0099] The text information creation prompt generator 1134 has an interface for accepting site-related information received from the user terminal 2 and prompt generation logic for generating a prompt for generating text information suitable for a website based on the received site-related information. The prompt generation logic has a function for selecting an optimal prompt template for generating text information suitable for a website. This selection uses conditional branching to select a prompt template that matches the site-related information and profile information.

[0100] The prompt templates for the text information creation prompt generation unit 1134 are prepared in advance. The prompt templates provide a framework for document structure and content that serves as the basis for the text information creation prompt generation unit 1134 to effectively create text information suitable for a website. These templates are customized and designed in advance according to the needs and circumstances of the user 4. Specifically, the prompt templates predefine questions and information formats tailored to the interests and intentions of the user 4. For example, if the user 4 is a lawyer and is creating a website for his or her law firm, the prompt template would include items such as a "catchphrase that emphasizes trustworthiness," a "detailed description of legal consultation services," and "an introduction of success stories." By using these prompt templates, the text information creation prompt generation unit 1134 can effectively extract specific text information that meets the needs of the user 4.

[0101] Furthermore, the text information creation prompt generator 1134 combines User 4's profile information and site-related information to grasp the overall picture of User 4's intended website and creates a prompt that generates text information in line with that intent. For example, if User 4 desires a "casual legal consultation website targeted at young people," the prompt template is customized to include content such as a "friendly catchphrase," "an explanation of legal consultation services for young people," and "a casual introduction to success stories." The prompt generation logic of the text information creation prompt generator 1134 selects the prompt template that best suits User 4's needs through complex conditional branching. In this process, more accurate prompts can be generated based on information learned from past data and similar cases. This prompt is sent to the large-scale language model unit 1131, generating the specific text information needed to create the website.

[0102] For example, catchphrases such as "A law firm that promises your trust and peace of mind" and "Legal experts support you" are generated. Descriptions of legal consultation services include "Our legal consultation services cover a wide range of areas, including contract drafting, business law, and labor issues. We provide advice and support tailored to each client's individual needs." and "Our firm offers free initial consultations. Please feel free to contact us." Examples of success stories include "We quickly resolved Company A's contract dispute and maintained the firm's credibility." and "We resolved Mr. B's labor issue and restored a fair working environment." Examples of FAQ enhancements include "Q: Is the initial consultation free?" and "Q: How do I schedule a legal consultation? A: You can easily schedule one by phone or through the website." Examples of profiles include "Taro Yamada: With over 20 years of experience as a lawyer, he specializes in business law and contract drafting. He has solved numerous legal problems for companies and individuals." and "Hanako Tanaka: As a UI / UX design expert, she has successfully led numerous projects. She also excels in branding." As characteristics of the service, "We have earned the trust of our customers through our quick response and thorough explanations," and "We provide individually customized legal advice" are generated.

[0103] This text information is customized to suit the intentions and needs of user 4 and used as website content. Furthermore, the text information generated by large-scale language model unit 1131 becomes more accurate based on specific information and prompts provided by user 4. This allows user 4 to efficiently create a website that best suits their needs.

[0104] The prompt transmission interface has the function of quickly and accurately transmitting the generated prompts to the large-scale language model unit 1131. The prompt transmission interface allows for smooth transmission and reception of prompts, enabling real-time information processing. This allows user 4 to efficiently create a website.

[0105] As described above, the text information creation prompt generation unit 1134 generates appropriate prompts based on the needs of user 4 and works in conjunction with the large-scale language model unit 1131 to provide web creation support information, thereby effectively supporting user 4 in creating a website.

[0106] (Website Creation Support Device 1: Analysis and Correction Function of Exit Rate) As shown in FIG. 6, it is preferable that the website creation support device 1 has the function of analyzing the dropout rate of each part of the website, identifying and pointing out parts that need improvement, and automatically suggesting corrections based on the points that are pointed out.

[0107] Specifically, the website creation support device 1 has an exit rate data collection unit 1161 that collects exit rates on websites on a web part basis, an evaluation information collection unit 1162 that collects evaluation information posted by visitors to the website, an exit rate analysis unit 1163 that analyzes the collected exit rates and identifies web parts with high exit rates, and a revision proposal creation support prompt generation unit 1164 that includes the web parts identified by the exit rate analysis unit 1163 in prompts of the large-scale language model unit 1131 and causes the large-scale language model to generate revision proposals and explanatory text for the identified web parts as web creation support information based on the evaluation information collected by the evaluation information collection unit 1162.

[0108] (Website creation support device 1: Exit rate analysis and correction function: Exit rate data collection unit 1161) The exit rate data collection unit 1161 has a function for collecting detailed information on the exit rate of visitors at each part of the website. Specifically, it is possible to clearly identify at which part of the website the visitor exited and which web part caused the exit. The exit rate data collection unit 1161 uses tracking technology to track how the visitor moved within the page and at what point they left the site.

[0109] The exit rate data collection unit 1161 monitors access data for each web part in real time, and records and saves that data in the exit rate database 1214 in the storage unit 12. This makes it possible to analyze the behavioral patterns of visitors to specific web parts, and to obtain basic data for identifying web parts with high exit rates. The exit rate data collection unit 1161 automatically collects data and updates it regularly, making it possible to perform analysis based on the latest information.

[0110] (Website creation support device 1: Absence rate analysis and correction function: Evaluation information collection unit 1162) The evaluation information collection unit 1162 has a function for collecting detailed evaluation information from visitors who visit a website. Specifically, it systematically records the evaluations and feedback provided by visitors to each web part and page. The evaluation information collection unit 1162 collects information directly entered by visitors through interfaces such as evaluation forms, feedback buttons, and questionnaires. The evaluation information collection unit 1162 collects and stores evaluation information in various formats, such as text-based comments and evaluation scores. This allows for a specific understanding of visitors' impressions of specific web parts and content and areas for improvement. This information is extremely important in evaluating website performance and serves as basic data for quantitatively analyzing high dropout rates and visitor satisfaction.

[0111] (Website Creation Support Device 1: Analysis and Correction Function of Exit Rate: Evaluation Database 1215) The evaluation information collected by the evaluation information collection unit 1162 is stored in the evaluation database 1215. The evaluation database 1215 is a database for systematically managing and storing evaluation information collected from website visitors. The evaluation database 1215 records in detail the evaluations and feedback given by visitors to each web part and page. The evaluation database 1215 includes evaluation information in various formats, such as text-based comments, evaluation scores, and survey response results, and this information serves as basic data for quantitatively analyzing visitor experiences and satisfaction.

[0112] The evaluation database 1215 includes the visitor ID, the page ID of the page that was evaluated, the web part ID of the web part that was evaluated, the evaluation score, which is the numerical score of the evaluation given by the visitor to the web part or page, text comments including detailed feedback and comments left by the visitor on the web part or page, and the survey results, which are the options and free-form comments given by the visitor when answering a survey.

[0113] The evaluation database 1215 is constructed using advanced database technology to efficiently store, search, and aggregate the above data. For example, by setting an index using a visitor ID or page ID as a key, it is possible to quickly search for evaluation information related to a specific visitor or page. It also incorporates text mining technology to analyze the evaluation scores and the contents of comments, enabling detailed analysis of visitors' impressions and opinions.

[0114] The information recorded in the evaluation database 1215 is used in conjunction with other components of the website creation support device 1, such as the exit rate analysis unit 1163 and the revision proposal creation support prompt generation unit 1164. For example, the exit rate analysis unit 1163 uses the information in the evaluation database 1215 to understand visitors' dissatisfaction with and areas for improvement for specific web parts, and identifies areas with high exit rates based on the results. The revision proposal creation support prompt generation unit 1164 uses the information in the evaluation database 1215 to propose specific revision proposals and improvement measures.

[0115] (Website creation support device 1: Exit rate analysis and correction function: Exit rate analysis unit 1163) The exit rate analysis unit 1163 has a function for analyzing the collected exit rate data in detail and identifying which parts of the website are causing visitors to exit. Specifically, it calculates the exit rate for each web part and page and clarifies which elements are causing problems for visitors. The exit rate analysis unit 1163 uses statistical methods and machine learning algorithms based on the collected data to analyze patterns and trends in the exit rate.

[0116] To explain in more detail how statistical methods are used to analyze patterns and trends in exit rates based on collected data, basic statistics are calculated. These include basic statistical indicators such as the average exit rate, median, and standard deviation. These indicators are used to understand the overall trend and variation in exit rates. Next, time series analysis is performed. Moving averages and models that take into account the effects of seasonality are used to analyze how exit rate data fluctuates over time. This identifies patterns of exit rate fluctuations over specific periods and evaluates the impact of seasons and events. Next, correlation analysis is performed. The correlation between exit rates between different web parts and pages is investigated to understand how specific elements affect other elements. For example, if a specific page has a high exit rate, analyzing the relationship between that page and other related pages and navigation elements can help identify the cause of the problem.

[0117] A segment analysis of the churn rate is also performed. Visitors are segmented based on demographic information and behavioral patterns, and the churn rate for each segment is analyzed. This makes it possible to clarify the tendency of certain visitor groups to churn under certain conditions and implement targeted improvement measures for that group. Finally, a regression analysis is performed. Multiple factors that affect the churn rate are identified, and the extent to which these factors affect the churn rate is quantitatively evaluated. This makes it possible to identify the most influential factors and prioritize improvement measures based on them.

[0118] Next, to explain in detail how a machine learning algorithm is used to analyze the patterns and trends of bounce rates based on collected data, various data related to bounce rates obtained during the data collection stage (e.g., visitor demographic information, access time, time spent on a page, click patterns, etc.) are preprocessed. Preprocessing includes data cleaning, filling in missing data, and detecting and processing outliers. Next, appropriate features are selected and input into the machine learning algorithm's learning model. When selecting features, variables that may affect the bounce rate are identified so that these variables optimize the model's performance. For example, factors such as page loading speed, type of content, and the visitor's past behavioral history are taken into consideration.

[0119] Next, a model is constructed using a machine learning algorithm (for example, decision tree, random forest, support vector machine, neural network, etc.). To train the model, the collected data is used as a training dataset so that the model can learn patterns and trends in churn rates. After the model is trained, the performance of the model is evaluated using a validation dataset. Evaluation metrics used include precision, recall, F1 score to balance precision and recall, and AUC-ROC (Area Under the Curve - Receiver Operating Characteristic) to evaluate the overall performance of the model. This allows us to confirm the degree of predictive accuracy of the model for actual data.

[0120] Next, the model is tuned to improve its performance through hyperparameter optimization. This process uses techniques such as cross-validation, grid search, and random search. Finally, the optimized model is used to analyze the patterns and trends of bounce rates in real time or in batches. The results of this analysis help identify the causes of high bounce rates in specific parts or pages of a website. For example, if the slow loading speed of a particular page is causing high bounce rates, specific improvements can be suggested, such as optimizing that page.

[0121] In addition, the exit rate analysis unit 1163 can utilize a variety of methods, such as heuristic analysis, visitor testing, heat map analysis, session replay, and feedback tools, in addition to statistical methods and machine learning algorithms, to analyze patterns and trends in exit rates based on the collected data.

[0122] Here, "heuristic analysis" is a method of problem-solving based on experience and intuition. This method forms hypotheses based on visitor behavior patterns and website characteristics to identify the causes of drop-off rates. For example, if a particular page has a high drop-off rate, problems with the page's design, content, and navigation can be intuitively evaluated. "Visitor testing" is a method of identifying the causes of drop-off rates by having actual visitors use the website and observing their behavior and reactions. Visitor testing makes it possible to directly observe drop-off points and causes of visitor frustration. This allows for a deeper understanding of the reasons why visitors leave.

[0123] "Heat map analysis" is a method that visually shows which parts of a web page visitors focus on and where they click. Using heat maps, you can understand which parts of the page visitors are abandoning and which elements are not attracting their attention. This allows you to identify the elements that cause abandonment and take measures to improve them. "Session replay" is a method that records visitors' behavior on a website and plays it back later to analyze the visitor's experience in detail. Using session replay, you can specifically observe how visitors operate a website and at what points they abandon it. This makes it possible to identify visitor behavior patterns and the causes of abandonment. "Feedback tools" are methods for collecting feedback directly from visitors. Using pop-up windows or feedback forms, you can ask visitors why they abandoned the website and what areas need improvement. Analyzing this feedback allows you to understand the causes of abandonment and take appropriate measures to improve it.

[0124] To improve the accuracy of the analysis, the churn rate analysis unit 1163 preferably uses a combination of the above-mentioned analytical methods. For example, when a statistical method is combined with a machine learning algorithm, the statistical method is effective in quickly grasping basic patterns and trends in data. For example, calculating basic statistical quantities such as the mean, median, and standard deviation allows for an intuitive understanding of the overall trend of the data. At this stage, outliers in the data can be identified, and understanding the basic distribution and range of variation can provide information useful for detailed analysis. On the other hand, machine learning algorithms excel at discovering complex patterns and nonlinear relationships that are often overlooked by statistical methods. For example, using algorithms such as decision trees, random forests, and neural networks, a predictive model can be constructed from the data, enabling more accurate predictions and classifications. This goes beyond simply grasping trends to identifying specific behavioral patterns and factors, enabling a detailed analysis of the churn rate.

[0125] Combining statistical methods with machine learning algorithms increases the breadth and depth of analysis. For example, statistical methods can be used to grasp the basic overview of data and identify specific patterns or anomalies. Then, by inputting that data into a machine learning algorithm, more advanced analysis can be performed. This process allows for quick understanding of overall data trends while revealing the complex factors behind the data through specific predictions and classifications. Furthermore, evaluating the results of machine learning algorithms with statistical methods makes it possible to verify the accuracy and reliability of the model. For example, model performance can be quantitatively evaluated by comparing the model's prediction results with actual data and calculating metrics such as mean error and standard deviation. This combination of statistical methods and machine learning algorithms is advantageous in that it quickly grasps basic data trends while also discovering complex patterns and nonlinear relationships, enabling highly accurate predictions and classifications. This combination improves the accuracy of churn rate analysis and enables effective improvement measures to be implemented.

[0126] The revision proposal creation support prompt generation unit 1164 has a role of including the web parts identified by the dropout rate analysis unit 1163 in the prompts of the large-scale language model unit 1131. Specifically, it has a function of generating revision proposals and explanations for the identified web parts using a large-scale language model based on the evaluation information collected by the evaluation information collection unit 1162. This revision proposal creation support prompt generation unit 1164 creates prompts to present improvement methods for web parts with high dropout rates.

[0127] The revision suggestion creation support prompt generation unit 1164 identifies web parts with high withdrawal rates using the withdrawal rate analysis unit 1163. Next, detailed information about the identified web parts is included in a prompt for the large-scale language model unit 1131. Furthermore, a prompt for generating specific revision suggestions is created based on visitor evaluation information and feedback collected by the evaluation information collection unit 1162. This prompt includes instructions for the large-scale language model to generate appropriate revision suggestions and accompanying explanatory text.

[0128] When generating the prompts, user 4's profile information and site-related information are also taken into consideration. This allows for the generation of revision suggestions that meet user 4's needs and expectations. For example, if visitor feedback indicates that navigation on a particular page is complicated, the revision suggestion creation support prompt generation unit 1164 generates a prompt containing specific suggestions for simplifying the navigation. This prompt is sent to the large-scale language model unit 1131, which generates specific revision suggestions and explanations. The generated revision suggestions and explanations are presented to user 4 to help improve the website.

[0129] (Website Creation Support Device 1: Analysis and Correction Function of Exit Rate: Processing Operation) The website creation support device 1 configured as described above has the functions of collecting visitor behavior data on a website, identifying areas with high dropout rates, and automatically generating improvement proposals. Specific processing operations of the website creation support device 1 will be explained below using the dropout rate database 1214 in FIG. 7.

[0130] The exit rate data collection unit 1161 records the behavior of visitors on each page of the website. For example, actions such as visitors clicking on the homepage or scrolling through the about page are recorded along with a timestamp (see "Exit Rate Data Table" in the figure). Data such as the visitor ID, page ID, action, and timestamp are recorded, which makes it possible to identify the exit behavior of visitors on each page.

[0131] The collected data is then sent to the exit rate analysis unit 1163, where it is analyzed using statistical methods and machine learning algorithms. This analysis calculates the exit rate for each web part (see "Exit Rate Analysis Data Table" in the figure). For example, specific values ​​are displayed, such as the exit rate for the homepage header being 0.05, the footer being 0.15, and the sidebar being 0.25. This makes it clear which web parts are causing problems for visitors.

[0132] A revision suggestion creation support prompt generation unit 1164 operates for web parts with high exit rates identified by the exit rate analysis unit 1163. Information about web parts with high exit rates is input as a prompt to the large-scale language model unit 1131, and revision suggestions and explanatory text are generated taking into consideration visitor evaluation information collected by the evaluation information collection unit 1162. The revision suggestions include specific suggestions on how to improve the web part (see the "Data table for the revision suggestion function" in the figure). For example, specific suggestions such as reducing clutter in the sidebar and adding contact information to the footer are provided.

[0133] The generated revision suggestions and explanations are then provided to the user terminal 2 as website creation support information, allowing the user 4 to make revisions to the website based on the specific improvement suggestions. Furthermore, the analysis results display dashboard displays the total number of visits, bounce rate, average time spent on page, and conversion rate in real time, making it possible to continuously monitor website performance.

[0134] (Website Creation Support Device 1: Exit Rate Analysis and Correction Function: Exit Rate Database 1214) The exit rate database 1214 records detailed data on the behavior of visitors to each page and web part on the website and stores data for identifying areas with high exit rates. Specifically, it has a "data table for exit rate data," a "data table for exit rate analysis," a "data table for the revision suggestion function," and a "data table for the analysis result display dashboard."

[0135] The "exit rate data table" is collected by the exit rate data collection unit 1161 and records the detailed behavioral history of each visitor who visits the website. This data table consists of four main items: visitor ID, page ID, action, and timestamp. The visitor ID uniquely identifies each visitor, and the page ID indicates the specific web page that the visitor visited. The action item records the specific action that the visitor took on that page (e.g., click, scroll, hover, exit, etc.). The timestamp indicates the exact date and time that these actions occurred and is used to track the chronological order of the actions.

[0136] For example, looking at the behavioral history of visitor ID1, we can see that they clicked on the homepage at 10:00 on January 1, 2023, and then performed a hover action on the same page at 10:15. Also, visitor ID2 is recorded as scrolling on the about page at 10:05, and then scrolling again on the same page at 10:20. Visitor ID3 is shown to have clicked on the contact page at 10:10, and then left that page at 10:25.

[0137] In this way, the "Bounce Rate Data Table" records in detail the specific actions and timing of each visitor, allowing you to understand which parts of your website are being used by visitors and how they are being used. Using this data, you can evaluate the performance of specific sections or content of your website and analyze visitor behavior patterns.

[0138] The "Exit Rate Analysis Data Table" records the exit rate of each web part in detail. This data table quantifies the extent to which each part of a website affects visitor exit, and is extremely important for identifying areas for improvement on a website. Specifically, it contains the web part ID, page ID, exit rate, and analysis timestamp. The web part ID is an identifier that uniquely identifies each web part, and the page ID indicates the page on which the web part is located. The exit rate is an indicator of the extent to which a particular web part contributes to visitor exit; a higher value means a greater impact as a cause of exit. The analysis timestamp indicates the date and time when this exit rate was measured and analyzed.

[0139] For example, the bounce rate for the homepage header is 0.05, which is a relatively low figure. This indicates that the header section has little impact on visitor bounces. On the other hand, the bounce rate for the banner on the contact page is high at 0.3, which indicates that visitors may have a negative reaction to the banner. Web parts with such high bounce rates are identified as areas that need improvement.

[0140] The exit rate analysis unit 1163 analyzes this data in real time and continuously evaluates the performance of the website. This allows for quantitative evaluation of the impact that specific web parts have on visitor exit and provides basic data for taking improvement measures. For example, for web parts with high exit rates, it becomes possible to propose specific improvement measures, such as reviewing the design or changing the content.

[0141] The "Revision Suggestion Function Data Table" records detailed revision suggestions for web parts with high dropout rates. This data serves as a basis for clarifying improvements needed for each part of a website and making efficient improvements. Specifically, it includes the web part ID, proposal content, priority, and proposal timestamp. The web part ID is an identifier that uniquely identifies each web part, allowing revision suggestions to be accurately associated with specific web parts. The proposal content details the specific revision suggestions and clearly indicates what kind of improvement measures are needed. The priority indicates the urgency and importance of the revision suggestion and is categorized as high, medium, or low. This allows for effective development of an improvement plan for the entire website. The proposal timestamp indicates the date and time the revision suggestion was suggested and is used to manage the revision suggestion history and check the progress of improvements.

[0142] For example, a low-priority suggestion for the homepage header, "Improve visibility," was recorded at 10:00 on January 3, 2023. This suggestion aims to redesign and position the header to make it easier for visitors to find information. Similarly, a high-priority suggestion for the sidebar, "Reduce clutter," was recorded, aiming to organize the sidebar elements so that visitors can easily access the information they need.

[0143] These specific suggested modifications allow for efficient website improvement. Implementing suggested modifications for each web part based on priority can reduce the bounce rate. Proposal timestamps also allow tracking of which suggested modifications were made and when, making it easy to manage the progress of improvement activities. Furthermore, the data from the suggested modifications function works in conjunction with the bounce rate analysis data. After identifying web parts with high bounce rates, suggested modifications are quickly generated and presented, shortening the website improvement cycle. This is expected to continuously improve the overall performance of the website and increase visitor satisfaction.

[0144] The "Analysis Results Display Dashboard Data Table" contains information for monitoring and evaluating the overall performance of a website in detail. This data table is an important source of information for website operators to understand the current status of their site and plan and implement specific actions for improvement.

[0145] The total number of visits indicates the total number of visitors to a website within a specific period of time, and is used to evaluate a website's popularity and ability to attract customers. For example, a total number of visits of 1,000 indicates that 1,000 visitors visited the website within the specified period. The exit rate indicates the percentage of visitors who left the site after viewing a specific page. For example, an exit rate of 0.2 means that 20% of visitors left the site after viewing a specific page. This data can be used to identify problematic pages or web parts and take measures to improve them.

[0146] Average time on page indicates the average time a visitor stays on a particular page and is used to evaluate the attractiveness of content and the ease of use of a page. For example, an average time on page of 2 minutes indicates that visitors stay on the page for an average of 2 minutes. This data is an indicator of how interested visitors are in the content. Conversion rate indicates the percentage of visitors who complete a targeted action on a website (e.g., purchase, registration, inquiry, etc.). This metric is important when evaluating the effectiveness of a website and business results. For example, a conversion rate of 5% means that 5% of visitors completed the targeted action. This data can be used to improve marketing strategies and conversion funnels.

[0147] The report timestamp indicates the date and time when each evaluation indicator was recorded, ensuring the freshness and real-time nature of the data. For example, a report timestamp of 10:00 on January 4, 2023 indicates that it is the latest performance data at that time. This allows website operators to understand the latest situation and respond quickly. In this way, the "Data Table on the Analysis Results Display Dashboard" provides information for evaluating the overall performance of a website from multiple perspectives.

[0148] (Website Creation Support Device 1: Analysis and Correction Function of Exit Rate: Effects) As described above, the website creation support device 1 includes an exit rate data collection unit 1161 that collects exit rates for websites on a web part basis, an evaluation information collection unit 1162 that collects evaluation information posted by visitors to the website, an exit rate analysis unit 1163 that analyzes the collected exit rates and identifies web parts with high exit rates, and a revision proposal creation support prompt generation unit 1164 that includes the web parts identified by the exit rate analysis unit 1163 in prompts of the large-scale language model unit 1131 and causes the large-scale language model to generate revision proposals and explanatory text for the identified web parts as web creation support information based on the evaluation information collected by the evaluation information collection unit 1162.

[0149] According to the above configuration, the exit rate data collection unit 1161 collects exit rate data for each web part, thereby collecting detailed behavioral data for each part of the website. This behavioral data allows for a clear understanding of which web parts are causing visitors to exit. Furthermore, the evaluation information collection unit 1162 collects evaluation information from visitors who visit the website. This evaluation information includes specific feedback and reviews from visitors, clarifying the problems and dissatisfaction they face. This information plays an important role in analyzing and improving the exit rate. The exit rate analysis unit 1163 analyzes the collected exit rate data in detail to identify web parts with high exit rates. This makes it possible to accurately identify which parts of the website are causing visitors to exit. This provides basic data for taking specific improvement measures for the identified problems. Furthermore, the revision proposal creation support prompt generation unit 1164 includes the identified web parts with high exit rates in prompts from the large-scale language model unit 1131 and generates revision proposals and explanations based on the evaluation information. By utilizing the large-scale language model, it is possible to analyze visitor feedback and reviews and automatically generate optimal revision proposals. These suggested revisions indicate specific improvements and provide guidelines for implementation. Additionally, by generating explanatory text, the background and purpose of the revisions can be clearly communicated. The revision suggestions and explanatory text generated in this way are provided as web development support information, enabling quick and effective improvements to websites. This is expected to reduce the bounce rate.

[0150] (Website Creation Support Device 1: Automatic Adjustment Function for Web Parts) It is preferable that the website creation support device 1 has a function for automatically adjusting colors and layouts when combining web parts, thereby maintaining a consistent design.

[0151] Specifically, as shown in FIG. 8, the website creation support device 1 has a design consistency maintenance unit 1171 that adjusts the color and layout of the website based on the theme color, brand guidelines, and existing design patterns contained in the site-related information when combining web parts; a style adjustment unit 1172 that adjusts the style of each web part based on the adjustment results of the design consistency maintenance unit 1171; and a design information communication control unit 1173 that transmits the website design information adjusted by the style adjustment unit 1172 to the user terminal 2 via the communication unit 13.

[0152] (Website creation support device 1: Web part automatic adjustment function: Design consistency maintenance unit 1171) The design consistency maintenance unit 1171 has an important function of automatically adjusting colors and layouts to maintain consistency in the design of the entire site when web parts are combined. That is, the design consistency maintenance unit 1171 has a function of maintaining the overall visual harmony of the website based on elements such as theme colors, brand guidelines, and existing design patterns contained in the site-related information.

[0153] Specifically, the design consistency maintenance unit 1171 analyzes site-related information and extracts basic color tones, font styles, and layout patterns based on theme colors and brand guidelines. This information is applied to ensure that web parts have an appropriately unified design. For example, if the brand guidelines specify a design based on blue, the design consistency maintenance unit 1171 adjusts the color scheme of all web parts to match that theme color. Adjustments are also made based on existing design patterns. Existing design patterns refer to successful designs and standard design templates from the past, and the website layout is automatically optimized based on these. For example, the placement of the navigation bar, button style, and text block placement are adjusted for consistency.

[0154] Furthermore, the design consistency maintenance unit 1171 also has a dynamic adjustment function. The dynamic adjustment function is a function that immediately responds when the user 4 adds or changes the placement of a web part, and performs processing to automatically maintain the consistency of the design. For example, when a new web part is added, the design consistency maintenance unit 1171 automatically adjusts the color, font style, and placement of the web part to match the existing theme color and brand guidelines. This ensures that the newly added part visually harmonizes with other parts, maintaining a unified design. The dynamic adjustment function is also applied when the placement of an existing web part is changed. For example, when the user 4 moves a part to a different position, the design elements around that part are automatically adjusted, maintaining a consistent layout overall. This includes adjusting margins, rearranging text, changing image sizes, etc.

[0155] Furthermore, the dynamic adjustment function provides real-time visual feedback each time User 4 adds or rearranges a web part. This allows User 4 to instantly see how the changes are reflected, enabling intuitive design adjustments. For example, when moving a part by dragging and dropping, User 4 can preview in real time what the part's new position will look like. In this way, the dynamic adjustment function enables User 4 to quickly build a website that has a unified overall look, without having to worry about adjusting the design of each individual part. This enables efficient website creation while maintaining design quality.

[0156] (Website creation support device 1: Automatic adjustment function of web parts: Style adjustment unit 1172) The style adjustment unit 1172 has a function of automatically adjusting the style of each web part based on the adjustment results of the design consistency maintenance unit 1171. This maintains the consistency and visual harmony of the design of the entire website.

[0157] Specifically, the style adjustment unit 1172 receives the adjustment results generated by the design consistency maintenance unit 1171 and sets the style of each web part based on the adjustment results. This process includes adjusting design elements such as color, font, margin, padding, layout, and size. For example, if the theme color of the entire website is blue, newly added web parts will also be automatically set to match that theme color.

[0158] The style adjustment unit 1172 ensures that each web part follows consistent design guidelines, maintaining a sense of visual unity. Even when user 4 customizes the style of the website, adjustments are made so as not to disrupt the overall design balance. For example, even if user 4 changes the font size of a specific part, other style elements are automatically adjusted to maintain overall harmony, taking into account the balance with other parts.

[0159] Furthermore, the style adjustment unit 1172 has a function of reflecting style changes in real time. This means that every time User 4 makes a change to the design, the change is immediately applied to the entire website, providing visual feedback. For example, if User 4 changes the style of a button, the change is immediately applied to the buttons on all pages, allowing User 4 to make adjustments while checking a consistent design.

[0160] In this way, the style adjustment unit 1172 automatically adjusts the style of each web part based on the adjustment results of the design consistency maintenance unit 1171, thereby maintaining consistency in the design of the entire website and providing a function to assist user 4 in creating an efficient and high-quality website.

[0161] (Website creation support device 1: Automatic adjustment function of web parts: Design information communication control unit 1173) The design information communication control unit 1173 has the function of quickly and accurately transmitting the website design information adjusted by the style adjustment unit 1172 to the user terminal 2. This allows the user 4 to check the design changes in real time and make further adjustments as necessary.

[0162] Specifically, the design information communication control unit 1173 has the function of receiving the design information generated by the style adjustment unit 1172 and transmitting it in an appropriate format to the user terminal 2. The design information includes style settings such as color, font, layout, and size of each web part. This information is immediately reflected in the web browser or design tool used by the user 4 and displayed as visual feedback.

[0163] The design information communication control unit 1173 uses a high-speed, reliable communication protocol to transmit design information with low latency. This allows user 4 to receive quick feedback because design changes are reflected in real time. For example, if user 4 changes the color of a button, the change is immediately reflected on all related pages, allowing user 4 to check the changed design and make further adjustments.

[0164] The design information communication control unit 1173 also has an error check function to ensure communication stability and data integrity when transmitting design information. This prevents data corruption and transmission errors, ensuring that accurate design information is always delivered to the user terminal 2. The error check function checks the integrity of the data before and after transmitting the design information, and retransmits the data if a problem occurs. The design information communication control unit 1173 also has a function to receive feedback from the user 4. When the user 4 requests changes or reports problems with the design, it collects that information and provides feedback to the style adjustment unit 1172 and the design integrity maintenance unit 1171 as necessary. In this way, the design information communication control unit 1173 quickly and accurately transmits the design information adjusted by the style adjustment unit 1172 to the user terminal 2, thereby supporting real-time design confirmation and adjustment and providing an important function that enables the user 4 to efficiently create high-quality websites.

[0165] The website creation support device 1 configured as described above helps the user 4 quickly create a unified design through an automatic adjustment function based on theme colors, brand guidelines, and existing design patterns, and further reduces the burden on the user 4 through detailed style optimization and real-time design confirmation functions, enabling efficient and high-quality website creation.

[0166] (Website Creation Support Device 1: Other Configurations) As shown in Figure 1, website creation support device 1 has website creation support unit 11, which is a computer, memory unit 12, and communication unit 13. Some or all of the components included in website creation support unit 11 may be configured as either hardware or software. Input to website creation support unit 11 is made via input unit 22, such as a keyboard, of user terminal 2 operated by user 4. In addition, web creation support information created by website creation support unit 11 is displayed on display unit 21, such as a display of user terminal 2.

[0167] The website creation support device 1 has an input device and a display device, both of which are not shown. Examples of the input device include a keyboard, a mouse, a touch panel, and a voice input device. Examples of the display device include a liquid crystal display device. This allows the website creation support device 1 to be configured using information processing devices such as general personal computers, laptop computers, smartphones, and tablet terminals. The website creation support device 1 may lack at least one of the input device and the display device. In this case, the input device and the display device are provided in an external terminal, both of which are not shown, and the website creation support device 1 can be used as a website creation support server.

[0168] In this embodiment, the functions of the website creation support device 1 are described as being installed in an information processing device, but the present invention is not limited to this, and the functions of the website creation support device 1 may be installed in cloud computing. In this case, cloud computing allows computer resources to be added as needed, making it possible to process large amounts of information, enabling faster and more efficient processing, and also making it possible to easily expand processing capacity to accommodate a significant increase in the number of users 4.

[0169] 5, the website creation support device 1 includes databases such as a profile database 1211, a web part management database 1212, a site-related database 1213, a dropout rate database 1214, and an evaluation database 1215. The website creation support device 1 may also include other databases 1216, such as a visitor behavior database that records detailed data on visitors' behavior on the website, a conversion database that records data on how website visitors converted (e.g., purchase, form submission, subscription registration), and an error / bug database that manages records of technical errors and bugs that occurred on the website. By combining these databases, the website creation support device 1 can collect and analyze data from various perspectives, such as visitor behavior, evaluations, and technical problems, and comprehensively improve the performance of the website.

[0170] (Website creation support program) The above has described the case where website creation support unit 11 of website creation support device 1 is equipped with functional blocks. Next, we will describe the case where these functional blocks are executed by processing steps of software (website creation support program). Note that each processing step (S1 to S14) in the website creation support program is executed in apparent parallel using multitasking control.

[0171] As shown in FIG. 9, the customer management process (S1) has the function of efficiently and securely managing all data related to the customer, i.e., user 4. This process is primarily performed using the profile database 1211 and includes a series of operations for registering, updating, deleting, and querying information about user 4. Specifically, a wide range of data is handled, from basic information about user 4 (such as name, address, and contact information) to detailed profile information (such as occupation, interests, and past website usage history). The customer management process (S1) involves the registration of a new customer. This process includes steps for entering basic customer information and generating a customer ID. Next, if information about an existing customer needs to be updated, an update process is performed to reflect the latest information. This update process is performed when the customer changes their own information or when the administrator modifies the data as needed.

[0172] Furthermore, the customer management process (S1) also performs customer authentication and authorization. This manages the access rights of each customer, who is a user 4, and ensures that only users 4 with appropriate permissions can access specific data and functions. Powerful security protocols are also implemented to prevent unauthorized access. The customer data collected and managed in this way is used to support website creation. For example, optimal web parts are selected and design suggestions are made based on the profile information of user 4. Personalized support is also provided by referencing user 4's past usage history.

[0173] For example, in the profile database 1211 in Figure 3, customer ID 1, Yamada Taro, is a 35-year-old man working as a lawyer at Yamada Law Office. He is interested in business law and contracts and has used the service three times. By managing information on customers like Yamada Taro, it is possible to provide him with the appropriate legal consultation services he needs. Customer ID 2, Tanaka Hanako, is a 28-year-old woman working as a designer at Tanaka Design Studio. She provides graphic design and is interested in UI / UX design and branding. She has used the service once, and based on this information, it is possible to provide optimal design services to Tanaka Hanako. Customer ID 3, Suzuki Ichiro, is a 45-year-old man working as an engineer at Suzuki Engineering. He is in charge of system development and is interested in AI and IoT. He has used the service five times in the past, and based on this data, optimal technical support and new development project proposals for Suzuki Ichiro are proposed.

[0174] In this way, the customer management process (S1) manages detailed customer profile information, making it possible to provide individually customized services to each customer.

[0175] The user interface process (S2) is a process for designing and managing the interface used by the user 4 when operating the website creation support device 1. This process is designed to allow the user 4 to operate it intuitively and efficiently. Specifically, the user interface process (S2) arranges and adjusts interface elements such as various input forms, buttons, menus, and navigation bars so that the user 4 can smoothly proceed through each step of website creation.

[0176] Specifically, interface elements such as various input forms, buttons, menus, and navigation bars are arranged and adjusted to allow user 4 to smoothly proceed through each step of website creation. For example, as shown in FIG. 10, the main category list in the leftmost area of ​​display unit 21 provides a menu for user 4 to select the type of web part they want to create. Categories displayed here include text, image, gallery, button, and form. User 4 can proceed to the next step by selecting the required category from this menu.

[0177] When User 4 selects the "Text" category, the "Text Category" menu appears. This menu includes specific text parts such as catchphrases, headings, paragraphs, lists, and quotes. User 4 can then select more specific text parts from here. Furthermore, if User 4 selects "Catchphrase" from the text category, the "Catchphrase List" appears, offering specific catchphrase options. User 4 can then select the most suitable catchphrase and add it to the website by dragging and dropping it into the editing window.

[0178] The menu list contents in the user interface process (S2) are designed to prioritize the display of optimal candidates from among the web part information selection candidates suitable for website creation generated by the website creation support unit 11. Specifically, the large-scale language model unit 1131 or the like proposes optimal web parts based on the user 4's profile information and site-related information, and these are reflected in the menu list. This mechanism allows user 4 to quickly find the parts that best suit their needs and purposes. For example, if user 4 is a lawyer and is creating a website for his or her law firm, the menu list will prioritize web parts such as a "catchphrase that emphasizes trustworthiness" and a "detailed explanation of legal consultation services." In this way, the user interface process (S2) supports the website creation process by efficiently providing web parts that match user 4's purposes.

[0179] In this way, the user interface process (S2) employs a hierarchical menu structure, allowing user 4 to easily find and select the desired web part. Furthermore, the drag-and-drop function allows for intuitive operation, allowing user 4 to easily add and edit the components of the website. Furthermore, the user interface process (S2) is equipped with a real-time feedback function. When user 4 adds or edits a web part, the results are displayed instantly, allowing user 4 to continue creating while checking the progress of their work. This feedback function allows user 4 to find the optimal design through trial and error.

[0180] The user input process (S3) is a process in which the user 4 inputs the information required for each step of creating a website. This process is responsible for accurately receiving the data entered by the user 4 and appropriately processing and saving it. Specifically, in the user input process (S3), the user 4 accesses the input form through the user interface and inputs the required information. For example, in the case of a text part, the user 4 can enter content such as a catchy slogan, heading, and paragraph. In the case of an image part, the user 4 uploads an image file and enters a description. At this time, the user 4 can instantly check the entered data through a real-time preview function displayed in the input form.

[0181] The data entered by User 4 is sent to the backend system, where it is converted into the appropriate format. Here, "backend system" refers to the entire system that operates behind the scenes of a website or application. Examples include database management and authentication and authorization. The backend system receives requests from the user interface (frontend), processes the data accordingly, and returns an appropriate response. The conversion process includes text formatting, image resizing, and data validation. Data validation verifies that the input is accurate and appropriate and checks for invalid data or formatting errors. For example, if a required field is left blank or the file format is not supported, an error message is displayed, prompting User 4 to re-enter the data. Once the conversion and validation are complete, the data is stored in the appropriate database in the storage unit 12. This storage process makes the entered data available for other processing steps and reflects the website's structure and design.

[0182] The web part management process (S4) manages the web parts that make up each part of a website. Specifically, it uses a web part management database 1212 to manage information on all available web parts. The web part management database 1212 records detailed information about each web part, such as the part ID, name, type, function, and creation date. By referring to the web part management database 1212, user 4 can easily find the web part they need.

[0183] The web part management process (S4) also utilizes a web part selection algorithm to select the web part that best suits the needs of user 4 and the requirements of the site. The web part selection algorithm analyzes the input information and profile information of user 4 and suggests the most suitable web part based on that information. Furthermore, the web part management process (S4) provides a web part management function specific to the customer (user 4). For example, it tracks web parts that a specific user 4 has used in the past or frequently used web parts, promoting reuse and improving work efficiency.

[0184] The web part management process (S4) also recommends the reuse of existing web parts by using a web part reuse algorithm. The web part reuse algorithm analyzes performance data and feedback from previously used web parts and re-proposes optimal parts based on that data. This allows user 4 to efficiently reuse high-quality web parts.

[0185] The web part management process (S4) may use a large-scale language model to automatically generate and customize web parts. In this case, natural language processing is used to generate web parts that meet the requirements of user 4, and to support the creation of more accurate websites, existing parts are improved and new parts are proposed.

[0186] The design generation process (S5) is a step for automatically generating the appearance and layout of a website. The purpose of this process is to integrate various design elements and provide a consistent design in order to realize the website design desired by user 4.

[0187] Specifically, the design generation process (S5) plans the overall design composition based on the web part information provided by the web part management process (S4). This includes all visual elements, such as the website layout, color scheme, font style, and image placement. These elements are assembled to create a consistent design in accordance with the user 4's requirements and brand guidelines. The design generation process (S5) also generates a design that is modified and adjusted to meet specific needs and requirements based on the information input by the user 4. For example, it uses machine learning algorithms and natural language processing technology to select and position design elements that match the user 4's desired design style and tone. It also uses image generation AI and text generation engines to generate unique visual and text content and incorporate it into the design.

[0188] The design generation process (S5) adjusts in accordance with theme colors and brand guidelines. This adjusts the overall color scheme and font style to fit User 4's brand image and design policies, maintaining consistency in the design. This achieves a professional, unified website design. The design generation process (S5) also provides a real-time feedback function. Every time User 4 adds or changes a design element, the preview is updated instantly. This feedback function allows User 4 to pursue the optimal design while checking in real time how the design changes will be reflected.

[0189] The automatic adjustment process (S6) is a step that dynamically optimizes the design elements and layout of the website. This process is performed to maintain consistency and visual harmony in the overall design when combining web parts.

[0190] Specifically, the automatic adjustment process (S6) automatically adjusts the colors and layout of the website based on the theme colors, brand guidelines, and existing design patterns contained in the site-related information. This eliminates the need for the user 4 to adjust design elements individually, making it easy to achieve a consistent appearance. For example, the hue, font size, and position of newly added web parts are dynamically adjusted so that they harmonize with the existing design.

[0191] The automatic adjustment process (S6) provides a function that maintains design consistency in real time even when User 4 adds a new web part or changes the placement of an existing web part. This allows User 4 to freely add and edit content while maintaining a sense of unity across the site without having to worry about adjusting the design. For example, if User 4 adds a "Contact Us" section, the design of that section is automatically adjusted to match the other sections.

[0192] Furthermore, the automatic adjustment process (S6) also makes adjustments to achieve a responsive design that takes into account the display on different devices and screen sizes used by the user 4. This makes it possible to provide consistent display content on a variety of devices, including smartphones, tablets, and desktop computers. For example, when viewing on a smartphone, the layout is converted to portrait orientation, and important information is automatically positioned for easy viewing.

[0193] The deployment and update process (S7) is a step that executes a series of procedures necessary to publish and maintain the generated and adjusted website. The purpose of this process is to ensure that the website works properly on the Internet 5 and that the contents are properly updated whenever users 4 make changes or updates.

[0194] Specifically, the deployment and update process (S7) uploads the completed website files and data to the designated web server 3 and makes it publicly available. This includes uploading static files such as HTML files, CSS files, JavaScript files, and image files, as well as setting up a database. When a user 4 clicks the publish button for the website, the deployment engine places these files and data in the appropriate folders, and the website becomes available on the designated domain.

[0195] The deployment and update process (S7) also automatically manages updates to existing websites. When User 4 changes the content of the website or adds new content, the deployment engine detects the changes, regenerates the necessary files, and uploads them to Web Server 3. This automatic update function ensures that User 4 always publishes the latest content, eliminating the need to manually upload files. The deployment and update process (S7) also manages versions of the changes. This is useful if User 4 wants to revert to a previous version or if a specific update causes a problem. The deployment engine records the change history of each version and can revert to a previous state if necessary. This function improves the stability and reliability of the website.

[0196] Furthermore, the deployment and update process (S7) monitors the performance and security of the published website. This includes monitoring the site's load time, error logs, and security threat detection. If the monitoring system detects an anomaly, a notification is sent to User 4, enabling a prompt response. This continuous monitoring helps maintain the website's performance and security, providing a high-quality service to User 4.

[0197] The customer-specific update notification process (S8) is a step for notifying each customer of updates and changes to the website in a timely manner. The purpose of this process is to ensure that updates made by users 4 are reflected and to promptly notify users 4 of important changes.

[0198] Specifically, the customer-specific update notification process (S8) notifies User 4 that changes to the website have been completed. This notification applies to all updates, including adding new website content, editing existing content, and changing the design. When User 4 saves the changes, the system automatically verifies the content and notifies User 4 that the changes have been successfully applied. Update notifications are modified and tailored to User 4's specific needs and requests based on User 4's settings. For example, User 4 can choose how often and how to receive notifications. Multiple methods are provided, including email notifications, SMS notifications, and push notifications, and can be configured according to User 4's preferences. This allows User 4 to receive important information in the way that suits them best.

[0199] The customer-specific update notification process (S8) also notifies users about important system updates and security patch applications. These notifications include system maintenance, the addition of new features, and the fixes for security vulnerabilities. By providing advance notification and encouraging users 4 to take necessary actions to prevent disruptions to website operations, system stability and security are maintained. Furthermore, these notifications also include feedback from users 4 about website performance. For example, if there is a change in key performance indicators, such as a sudden increase in the number of visitors or an abnormal increase in the bounce rate, this information is notified to users 4. The customer-specific update notification process (S8) also has a function for managing the history of notification content. Users 4 can review past notifications and recheck update history and important messages. This history management function allows users 4 to refer to necessary information at any time, helping them manage their websites.

[0200] The customer-specific update reflection process (S9) is a step for appropriately reflecting updates and changes made to the website of each user 4. The purpose of this process is to quickly and accurately apply the requests and changes of user 4 to the website.

[0201] Specifically, the Customer Update Reflection Process (S9) initiates the process of reflecting the updates and changes made by User 4 on the website. This includes any type of change, such as adding or editing text or images, adjusting the design, or adding features. Once User 4 confirms the changes, the content is analyzed and the procedure for applying the updates is automatically determined. The updates are then reflected on each web page. This includes placing the changed content in the appropriate position and applying the updates while maintaining consistency with existing content. For example, if new text is added, the text is adjusted so that it is positioned balanced with other content. Furthermore, if there are design changes, the overall layout and style are updated to maintain consistency.

[0202] The customer-specific update reflection process (S9) also includes a verification step to confirm that the updates have been reflected correctly. This verification step checks whether the updates are displayed as intended and whether the functions are working correctly. For example, if a new function is added, it tests whether that function works properly and checks for any defects. It also includes a function to clear the cache after the updates have been reflected. This allows users 4 and visitors to immediately check the latest information. Clearing the cache is especially important when large-scale changes have been made, as it prevents old data from being displayed.

[0203] The customer-specific update reflection process (S9) also has a function for managing the update history, which allows the user 4 to check the update details and change history that have been made in the past. This history management function is useful when it is necessary to return to a previous state or when past changes are needed as a reference. Furthermore, the customer-specific update reflection process (S9) also has a function for notifying the user 4 that the update has been completed.

[0204] The text management process (S10) is a step for creating, editing, saving, and managing text content used on a website. The purpose of this process is to help the user 4 efficiently manage the text content on the website and place it in an appropriate location.

[0205] Specifically, the text management process (S10) collects text data entered by the user 4 and edits or corrects it as necessary. This includes creating text elements such as headings, paragraphs, lists, and quotations. The user 4 can easily enter text and adjust the format using a text editor. The text management process (S10) also has a version management function that automatically saves the entered text and allows it to be restored to a previous version if necessary.

[0206] Furthermore, the text management process (S10) also provides functions for maintaining the consistency and quality of text content across the entire website. For example, it performs spelling checks, grammar checks, and applies style guides to ensure text quality. It also supports optimal keyword placement and metadata editing from an SEO (search engine optimization) perspective. This improves the website's search engine ranking and increases visitor access. The text management process (S10) works in conjunction with a text database. This database stores all text content, allowing users 4 to quickly search and access the text they need.

[0207] The exit rate data collection process (S11) is a step for collecting and analyzing detailed visitor exit rates for each part of the website. This process plays an important role in evaluating the performance of the website and identifying areas that need improvement.

[0208] Specifically, the exit rate data collection process (S11) tracks behavioral data of visitors in real time when they visit a website. This data includes the timing of page transitions, the time spent on each page, where specific actions (clicks, scrolls, hover, etc.) were performed, and the point at which the visitor left the site. This data is automatically collected and stored in the exit rate database 1214.

[0209] The exit rate data collection process (S11) uses tracking technology to record visitor behavior in detail. Specifically, a unique visitor ID is assigned to each visitor, and the pages visited on the site and the actions performed by that visitor are tracked. Page IDs and action timestamps are also recorded, making it possible to identify the specific timing and location of exit. For example, if visitor ID 1 clicks on the homepage and then leaves the site after staying there for a certain amount of time, a detailed behavioral history is saved as data.

[0210] Furthermore, the exit rate data collection process (S11) also has the function of monitoring access data for each web part in real time. This provides basic data for analyzing the extent to which a specific web part affects visitor exit. The collected data is used to identify web parts with high exit rates.

[0211] The exit rate analysis process (S12) is a step in which the patterns and trends of exit rates in each part of the website are analyzed based on the collected visitor behavior data. This process plays an important role in improving website performance and increasing visitor satisfaction.

[0212] Specifically, the exit rate analysis process (S12) uses the data collected in the exit rate data collection process (S11) to calculate the exit rate for each web part and analyzes the data in detail. The exit rate is an index that indicates the rate at which visitors leave the site at a specific web part or page, and this index can be used to identify which parts need improvement.

[0213] The exit rate analysis process (S12) uses a combination of statistical methods and machine learning algorithms. Statistical methods are effective in quickly identifying basic patterns and trends in data. For example, basic statistical quantities such as the mean, median, and standard deviation are calculated to understand the overall trend of the data. This makes it possible to quickly identify pages and web parts with high exit rates. On the other hand, machine learning algorithms are excellent at discovering more complex patterns and non-linear relationships. For example, by using algorithms such as decision trees, random forests, and neural networks, it is possible to build predictive models from the data, enabling more accurate predictions and classifications. This makes it possible to identify the causes of high exit rates and obtain analysis results that can be used to propose improvement measures.

[0214] Furthermore, the exit rate analysis process (S12) utilizes a variety of techniques, including heuristic analysis, customer testing, heat map analysis, session replay, and feedback tools. Combining these techniques enables a more comprehensive analysis and improves the accuracy of the analysis. For example, heat map analysis can visually show which parts of a page visitors are focusing on, and session replay can reproduce the specific behavior of visitors for detailed analysis.

[0215] The proposed correction process (S13) is a step in which specific proposed corrections for the identified problems are presented to the user 4 based on the results of the dropout rate analysis process (S12). The purpose of this process is to support the user 4 in quickly implementing effective improvement measures to improve website performance.

[0216] Specifically, the modification proposal presentation process (S13) generates improvement proposals for the web parts and pages with high bounce rates identified in the bounce rate analysis process (S12) and displays them to user 4. These proposals include a wide range of suggestions, such as changing the design or content of web parts, or adding or removing functions. The modification proposals are provided with detailed explanations so that user 4 can easily understand and put them into practice.

[0217] The revision proposal presentation process (S13) has a function of presenting the revision proposal generated by the revision proposal creation support prompt generation unit 1164 to the user 4 in a visually easy-to-understand manner. Specifically, the contents of the revision proposal are displayed as a pop-up window, a notification banner, or an alert on the dashboard. This allows the user 4 to quickly understand which parts of the website need improvement and take appropriate action.

[0218] The proposed revision process (S13) can change its display method depending on the priority and urgency of the proposed revision. For example, a proposed revision with a high priority requires immediate action and is therefore displayed prominently in the center of the screen. On the other hand, a proposed revision with a low priority is displayed in the notification center or as part of the dashboard, so that the user 4 can check it later. In this way, by adopting a display method according to the importance of the proposed revision, the burden on the user 4 is reduced and effective action is promoted. Furthermore, the proposed revision process (S13) has a function to track the implementation status of the proposed revision. When the user 4 takes action based on the proposed revision, the results are recorded and used for later analysis. This makes it possible to evaluate whether the proposed revision actually contributed to improving the website's performance. This feedback loop continuously verifies the effectiveness of the proposed revision and accumulates data for making more accurate proposals.

[0219] The analysis result display process (S14) is a step for visually displaying the website performance and behavioral data of user 4. The purpose of this process is to analyze the collected data and enable user 4 to intuitively grasp the current state of the website.

[0220] Specifically, the analysis result display process (S14) aggregates key evaluation indicators such as the exit rate, number of visitors, duration of stay, and conversion rate, and displays them in the form of a dashboard or report. This allows user 4 to monitor website performance in real time and quickly implement necessary improvement measures. The analysis result display process (S14) obtains necessary data from various databases and generates graphs and charts based on that data. For example, based on data obtained from the exit rate database 1214, the transition of the exit rate for each page is displayed in a line graph. In addition, data on the conversion rate and number of visitors is visually represented in bar graphs and pie charts. This allows user 4 to understand the trends and fluctuations in the data at a glance.

[0221] Furthermore, the analysis result display process (S14) has a function that allows user 4 to view detailed data through a user interface. User 4 can display detailed analysis results narrowed down to a specific period or specific page. This function allows user 4 to more specifically identify the cause of a problem and obtain information to take appropriate measures. The analysis result display process (S14) provides real-time data updates and good operability, allowing user 4 to make quick decisions based on the data. For example, if there is a sudden change in the number of visitors or conversion rate, user 4 can immediately analyze the cause and take appropriate measures. Such real-time data display and operability make it possible to constantly optimize website performance.

[0222] The analysis result display process (S14) also allows comparison with past data and trend analysis. This allows user 4 to understand how website performance is changing over time and provides basic information for considering long-term improvement measures. This trend analysis is also useful for evaluating seasonal fluctuations and the effectiveness of specific campaigns.

[0223] The above-described website creation assistance program can be installed in an information processing device such as a personal computer or tablet terminal to cause the information processing device to function as website creation assistance device 1. The program may be distributed in a state recorded on a computer-readable recording medium such as a CD-ROM or USB memory, or may be distributed via a two-way communication network such as the Internet or a one-way communication network or communication line such as television broadcasting. In other words, initial setting storage unit 16 may be a computer-readable recording medium such as a CD-ROM or USB memory.

[0224] (Website creation support method) The website creation support device 1 is configured to cause a computer (website creation support unit 11) to execute a website creation support method. Specifically, this is a method for causing the computer (website creation support unit 11) to execute each of the above processing steps.

[0225] It should be noted that within the scope of the concept of the present invention, those skilled in the art may conceive of various modifications and alterations. Therefore, it is understood that such modifications and alterations fall within the scope of the present invention. For example, those skilled in the art may appropriately add, delete, or modify components of the above-described embodiments, or may add, omit, or change the conditions of processes, as long as they maintain the essence of the present invention. [Explanation of symbols]

[0226] 1. Website creation support device 2. User terminal 3 Web Server 4 User 5. Internet 11 Website Creation Support Department 12 Storage section 13 Communications Department 1211 Profile Database 1212 Web Part Management Database 1213 Site Related Database 1214 Abandonment Rate Database 1215 Rating Database 1131 Large-scale Language Modeling 1132 Selection candidate creation prompt generation unit 1133 Proposal information creation prompt generation unit 1134 Text Information Creation Prompt Generation Unit 1161 Withdrawal Rate Data Collection Department 1162 Evaluation Information Collection Department 1163 Withdrawal Rate Analysis Department 1164 Revision Proposal Creation Support Prompt Generation Unit 1171 Design Integrity Maintenance Department 1172 Style Adjustment Section 1173 Design Information and Communication Control Unit

Claims

1. a web part management database in which information on web parts, which are elements that make up a website, is stored; a profile database in which profile information of users who create the website is stored; a communication unit that enables communication of site-related information about the website with a user terminal operated by the user; a website creation support unit that predicts the website desired by the user based on the profile information and the site-related information, extracts from the web part management database selection candidates for web part information suitable for creating the website, and presents the selection candidates to the user as web creation support information to support the creation of the website. Website creation support device.

2. The website creation support unit a large-scale language model unit having a large-scale language model that probabilistically predicts how likely words and sentences given in the prompt are to occur in natural language by combining one or more types of natural language processing; a selection candidate creation prompt generation unit that, when receiving the site-related information from the user terminal through the communication unit, includes the site-related information in the prompt of the large-scale language model unit, predicts the website desired by the user based on the web part information of the web part management database and the profile information of the profile database, and causes the large-scale language model to generate selection candidates for the web part information suitable for creating the website as the web creation support information.

2. The website creation support device according to claim 1.

3. The website creation support unit When the selection candidates of the web part information are generated, a suggested information creation prompt generation unit that includes the site-related information in the prompt of the large-scale language model unit and causes the large-scale language model to generate suggested information indicating a combination of selection candidates of the web part information that will become the website preferred by the user as the web creation support information based on the profile information of the profile database; 3. The website creation support device according to claim 2.

4. When the selection candidates of the web part information are generated, a text information creation prompt generation unit that includes the site-related information in the prompt of the large-scale language model unit and causes the large-scale language model to generate text information suitable for the website as the web creation support information based on the profile information in the profile database; 3. The website creation support device according to claim 2.

5. a dropout rate data collection unit that collects dropout rates on the website for each of the web parts; an evaluation information collection unit that collects evaluation information submitted by visitors to the website; a dropout rate analysis unit that analyzes the collected dropout rates and identifies the web parts with high dropout rates; a revision proposal creation support prompt generation unit that includes the web parts identified by the dropout rate analysis unit in the prompt of the large-scale language model unit, and causes the large-scale language model to generate revision proposals and explanations for the identified web parts as the web creation support information based on the evaluation information collected by the evaluation information collection unit.

3. The website creation support device according to claim 2.

6. a design consistency maintenance unit that adjusts the color and layout of the website based on a theme color, brand guidelines, and existing design patterns included in the site-related information when the web parts are combined; a style adjustment unit that adjusts the style of each web part based on the adjustment result of the design consistency maintenance unit; a design information communication control unit that transmits the design information of the website adjusted by the style adjustment unit to the user terminal via the communication unit.

2. The website creation support device according to claim 1.

Images