Automatic webpage content generation system

Abstract

The present invention provides a web page content automatic generation system that can be easily produced without editing a programming language. The web page content automatic generation system comprises: an image data acquisition unit that acquires web page content as an image data; a source code acquisition unit that acquires source code represented by HTML data constituting the web page content; a learning unit that acquires teaching data by analyzing the image data and the source code; an image data registration unit that inputs the web page content to be produced as an image data; an analysis unit that analyzes the correlation between the input image data and the teaching data stored in the teaching data storage unit; and a content source code generation unit that generates content source code related to the input image data based on the source code of the teaching data with a high correlation analyzed by the analysis unit.

Description

Technical Field

[0001] The present invention relates to a web content automatic generation system, and in particular to a web content automatic generation system useful as a system for automatically generating web pages published on the Internet. Background Art

[0002] In recent years, there has been a sharp increase in the number of companies or operators creating web pages that introduce their products and services and publicly publishing them as websites on the Internet in order to sell their own products and services. Consumers can browse the websites of various companies published on the Internet and use them as references when purchasing products and services.

[0003] Such websites are created using markup languages ​​such as HTML (Hyper Text Markup Language), CSS (Cascading Style Sheet), a language for specifying the style of web pages, programming languages ​​such as JS (Java Script (registered trademark)), which enables web pages to be actionable and interactive, and image data in data form such as jpg or png for displaying photos or illustrations. When creating and publishing a website, knowledge of the markup language and programming language used to create content, as well as knowledge of web servers used to upload the created content to the web servers, is required. These operations are usually performed by web page creators who have knowledge and experience.

[0004] However, it is said that there are billions of such websites around the world. Once a website is created, it can remain public as long as the information disclosed does not expire. Therefore, the number of public websites continues to increase.

[0005] Furthermore, as the number of websites increases, the importance of having a website for operators is increasing. People in industries that have never had a website or those that were considered unrelated to websites are starting to create websites. As a result, the demand for website creation is increasing, the recruitment rate for web page creators is increasing, and the supply is insufficient due to the shortage of human resources.

[0006] In addition, not only the number of websites has increased, but also the versions of the above-mentioned languages ​​such as HTML, CSS and JS have been improved, so the level required when creating websites has increased, which has become a cause of increasing the burden on web page creators.

[0007] As a technology for reducing the burden of website creation, there is a method using a template as shown in Patent Document 1 (Patent Document 1). According to the technology described in Patent Document 1, materials required for website creation, such as articles and images to be published on the website, are prepared, and these materials are applied to a predetermined framework such as a template, thereby making it easy to create a website even without professional programming knowledge.

[0008] Prior Art Literature

[0009] Patent Literature

[0010] Patent Document 1: Japanese Patent Application No. 2016-519347 Summary of the invention

[0011] Problems to be solved by the present invention

[0012] By utilizing such technology, people who wish to create a website can create the website without actually using a programming language, thereby achieving an effect of greatly saving the time required to create the website.

[0013] However, in a common website, a plurality of articles, images and / or videos are arranged in a complicated manner. When the method for creating a website using a template described in the above-mentioned patent document 1 is used, it takes time to apply each article, image and / or video to each part of the pre-prepared template, resulting in a problem of poor work efficiency.

[0014] In addition, if it is only applied to the template, the overall balance of the page will be deteriorated, and after application, it may be necessary to make fine adjustments in the arrangement of articles, images, etc., the number of characters per line, the position of line breaks, etc. In this case, the fine adjustments are made by editing the markup language and programming language, which further deteriorates the work efficiency.

[0015] Furthermore, the template prepared in advance is usually a template selected by the user from a plurality of templates, so the above-mentioned fine adjustment is different depending on the type of the selected template and becomes a necessary operation every time a website is created. Therefore, further deterioration of the operating efficiency is inevitable, and it is desired to realize a website creation system that reduces the burden on the operator.

[0016] The present invention has been made in view of such a desire, and an object of the present invention is to provide a system for automatically generating web page contents that can easily create even a website having a complicated structure without editing using a programming language.

[0017] Means used to solve problems

[0018] A web page content automatic generation system of one scheme is a web page content automatic generation system that automatically generates web page content described in HTML, wherein the system comprises: an image data acquisition unit that acquires web page content posted on a website open to the Internet as image data; a source code acquisition unit that acquires source code represented by HTML data constituting the web page content; a learning unit that analyzes the image data acquired by the image data acquisition unit and the source code acquired by the source code acquisition unit to thereby acquire teaching data; an image data registration unit that inputs web page content to be created as image data; an analysis unit that analyzes the correlation between the input image data and the teaching data acquired by the learning unit; and a content source code generation unit that generates content source code related to the input image data based on the HTML data of the teaching data with a high correlation analyzed by the analysis unit.

[0019] According to a web page content automatic generation system of one aspect, even a website with a complex structure can be easily created without editing a programming language.

[0020] The web content automatic generation system of one aspect may further include a website search unit for searching websites, wherein the website search unit automatically searches for websites published on the Internet using a search engine based on input search information.

[0021] According to an automatic web page content generation system of one aspect, by providing a search engine that automatically searches for websites and collects teaching data related to the searched websites, not only the creation of websites but also the collection of teaching data can be automated.

[0022] A web page content automatic generation system of one scheme comprises: a labeling unit, which obtains web page content posted on a website open to the Internet and labels the web page content according to each content; a production unit, which produces a structure that constitutes a part of the web page content based on the source code used to generate the web page content; a learning unit, which obtains teaching data based on the content labeled by the labeling unit and based on the source code and the structure produced by the production unit; a web page design acquisition unit, which obtains a web page design for producing the web page content; and a content source code generation unit, which obtains the content of the web page design based on the web page design obtained by the web page design acquisition unit and the teaching data obtained by the learning unit, and generates source code corresponding to the content.

[0023] In the web page content automatic generation system according to one aspect, when there is content corresponding to a component in the web page design content, the content source code generation unit may generate a source code corresponding to the component based on the teaching data.

[0024] In the web page content automatic generation system according to one aspect, the content source code generation unit may generate position information for arranging the content in the web page content based on the position of the content described in the web page design.

[0025] In one scheme of a web page content automatic generation system, the web page design acquisition unit may acquire image data corresponding to an image when the web page design includes an image, and the content source code generation unit may generate position information for configuring an image based on the image data within the web page content based on the position of the image within the web page design.

[0026] A web page content automatic generation system of one scheme comprises: a labeling unit, which obtains web page content posted on a website open to the Internet and labels the web page content according to each content; a production unit, which produces a structure that constitutes a part of the web page content based on the source code used to generate the web page content; a learning unit, which obtains teaching data based on the content labeled by the labeling unit and based on the source code and the structure produced by the production unit; a web page design acquisition unit, which obtains a web page design for producing the web page content; and a content source code generation unit, which obtains the content of the web page design based on the web page design obtained by the web page design acquisition unit and the teaching data obtained by the learning unit, and generates source code corresponding to the content.

[0027] In the web page content automation system according to one aspect, when a part of the content of the web page design corresponds to the structure, the content source code generating unit may generate the source code corresponding to the structure based on the teaching data.

[0028] In one aspect of the web page content automatic generation system, the content source code generation unit may generate position information related to a position where the content in the web page content is arranged based on the position of the content in the web page design so as to correspond to the layout of the content in the web page design.

[0029] In one scheme of a web page content automatic generation system, the web page design acquisition unit may acquire image data corresponding to an image when the web page design includes an image, and the content source code generation unit may generate position information for configuring an image based on the image data within the web page content based on the position of the image within the web page design.

[0030] Effects of the Invention

[0031] According to the web page content automatic generation system, even a website with a complex structure can be easily created without editing the markup language and programming language. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1It is a block diagram showing the hardware structure and software functions of the web content automatic generation system 1 in the first embodiment.

[0033] Figure 2 This is a flowchart showing the preliminary preparation for automatic generation of web page contents.

[0034] Figure 3 This is a diagram showing an example of the web page database 310 .

[0035] Figure 4 1 is a diagram showing an example of image data 1000 obtained by imaging a web page.

[0036] Figure 5 This is a diagram showing an example of HTML data 2000 which is a source code describing a web page.

[0037] Figure 6 This is a flowchart showing a method for creating web page contents.

[0038] Figure 7 This is a block diagram for explaining the web content automatic generation system according to the second embodiment.

[0039] Figure 8 This is a flowchart for explaining the flow of processing when creating teaching data according to the second embodiment.

[0040] Fig. 9 This is a flowchart for explaining the flow of processing when a web page is automatically generated according to the second embodiment.

[0041] Fig.10 This is a block diagram for explaining a web page content automatic generation system according to a modified example. DETAILED DESCRIPTION

[0042] Hereinafter, the mode for implementing the present invention will be described with reference to the drawings. It should be noted that this is just an example, and the technical scope of the present invention is not limited to this.

[0043] [First embodiment]

[0044] First, the first embodiment will be described.

[0045] [Structure of the web page content automatic generation system 1]

[0046] Figure 1 This is a block diagram for explaining the hardware configuration and software functions of the web content automatic generation system 1 in the first embodiment.

[0047] The web content automatic generation system 1 includes a control unit 100 for controlling data, a communication unit 200 for communicating with a user and other devices, a storage unit 300 for storing data, an input unit 400 for accepting input of information from a user, and a display unit 500 for outputting data and images controlled by the control unit 100.

[0048] The control unit 100 includes a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), a GPU (Graphic Processing Unit), and the like.

[0049] The communication unit 200 includes a device for communicating with other devices, for example, a Wi-Fi (Wireless FIdelity) compatible device in accordance with IEEE802.11 and a device equipped with hardware capable of utilizing the TCP / IP protocol.

[0050] The control unit 100 reads the specified program and cooperates with the communication unit 200 and the storage unit 300 as needed, thereby realizing the retrieval information registration module 111, the website retrieval module 112, the web page display module 113, the image data acquisition module 114, the HTML data acquisition module (source code acquisition module) 115, the learning module 116, the teaching data storage module 117, the image data registration module 118, the image analysis module 119, and the content source code generation module 120.

[0051] The storage unit 300 is a device for storing data and files, and includes a data storage unit composed of a hard disk, a semiconductor memory, a recording medium, a memory card, etc. The storage unit 300 stores a web page database 310, a teaching data database 320, and a web page content database 330 described later.

[0052] There is no particular limitation on the type of the input unit 400. Examples of the input unit 400 include a keyboard, a mouse, and a touch panel.

[0053] There is no particular limitation on the type of the display unit 500. Examples of the display unit 500 include a monitor and a touch panel.

[0054] [Preliminary preparation using the web content automatic generation system 1]

[0055] The following describes the preparations to be performed before executing content creation using the content automatic generation system 1 according to the present embodiment. Figure 2 This is a flowchart prepared in advance using the web content automatic generation system 1. In addition, Figure 3 This is a diagram showing an example of the web page database 310 . Figure 4 1 is a diagram showing an example of image data 1000 obtained by imaging a web page. Figure 5 2 is a diagram showing an example of HTML data 2000 which is one of the source codes describing a web page. Figure 1 to Figure 5 , the processing performed by the above-mentioned hardware and software modules is explained.

[0056] [Steps S110 to S190: Collection of information related to the web page]

[0057] First, as a preliminary preparation for implementing web page content creation, information related to the associated web pages is collected. A user (including an operator, hereinafter the same) who wants to use the web page content automatic generation system 1 to create web page content uses a user terminal (including an operator terminal, hereinafter the same) to use, inputs search information for searching websites that publish web pages to be collected as teaching data in a prescribed format (step S110), accesses the web page content automatic generation system 1 via the Internet, and sends search information for searching websites to be collected (step S120).

[0058] Here, as the search information input, for example, in the case of wanting to create a mail-order sales site for fashion for professional women, keywords such as "professional", "female", "fashion", and "mail-order sales" are cited for searching related websites. It should be noted that in step S120, the search information is sent from the user terminal to the web page content automatic generation system 1 via the Internet, but the web page content automatic generation system 1 can also be integrated with the user terminal.

[0059] When receiving the search information from the user terminal, the control unit 100 and the storage unit 300 of the webpage content automatic generation system 1 cooperate to execute the search information registration module 111 to register the search information input by the user operation in the Figure 3 The web page database 310 of the storage unit 300 is shown (step S130).

[0060] exist Figure 3 The web page database 310 shown stores the search information registered in step S130, the URL (Uniform Resource Locator) associated with the search information and assigned to the collected web page described later, the image data obtained by visualizing the web page, the image data displayed on the web page, the HTML data that describes the source code of the web page, the CSS data for specifying the style of the web page, and the JS data for making the web page actionable and interactive, wherein the search information matches the keywords used to search the website that publishes the web page to be collected as teaching data.

[0061] Next, the control unit 100 and the communication unit 200 of the web page content automatic generation system 1 cooperate to execute the website search module 112 to access the Internet and search for websites matching the keywords based on the keywords registered in the web page database 310 (step S140).

[0062] When the list of search results is obtained in step S140 , the control unit 100 and the display unit 500 of the web content automatic generation system 1 cooperate to execute the web page display module 113 to browse the websites searched in step S140 on the browser (step S150 ).

[0063] Then, the control unit 100 of the web page content automatic generation system 1 executes the image data acquisition module 114 to obtain the screen of the web page published on the website as the image data 1000 (step S160), and then cooperates with the storage unit 300 to associate the obtained image data 1000 with the URL of the web page and store it in the web page database 310 (step S170).

[0064] When the screen displayed as a web page is visualized, the screen displayed on the browser is scanned end-to-end and connected together as needed, thereby obtaining a single visualization data 1000. The obtained visualization data 1000 is, for example, Figure 4 Shown in.

[0065] exist Figure 4 In the image data, images related to commodities and images representing activities are arranged, and texts serving as annotations are given to each image. In step S160, all such texts and images are obtained as one image data.

[0066] At the same time, the control unit 100 of the web page content automatic generation system 1 executes the source code acquisition module 115 to obtain the source code of the browsed website composed of HTML data 2000, CSS data and JS data as text data (step S180), and then cooperates with the storage unit 300 to associate the obtained source code with the URL of the web page and store it in the web page database 310 (step S190).

[0067] The source code can be obtained on the browser, and the HTML data 2000 in the obtained source code is, for example, Figure 5 Shown in.

[0068] Such storage of the image data 1000 and the source code in the web page database 310 is repeatedly performed for the website searched in step S140 , thereby constructing the web page database 310 .

[0069] It should be noted that in the steps from searching for a website in step S140 to collecting source code in step S190 , software called a search engine automatically searches for a website and repeatedly browses web pages, thereby automatically collecting a plurality of image data 1000 and source code.

[0070] On the other hand, instead of using the website search module 112 to automatically search for a list of websites for collecting data, the list may be provided by a user or an operator.

[0071] [Steps S200 to S210: Associating the image data with the source code for learning]

[0072] When the construction of the web page database 310 is completed, the control unit 100 of the web page content automatic generation system 1 executes the learning module 116, and learns based on the search information stored in step S130, the image data 1000 stored in step S170, and the source code stored in step S190 to generate teaching data 3000 (step S200). For example, by using a machine learning model such as a convolutional neural network (CNN), the information related to the content and configuration of the image and article is associated with the source code to perform the learning. It should be noted that the learning in step S200 can be performed after collecting information related to all web pages searched in the retrieval of step S140, or it can collect information related to one web page and perform learning each time.

[0073] Then, the control unit 100 and the storage unit 300 of the web content automatic generation system 1 cooperate with each other to execute the teaching data storage module 117 to store the teaching data 3000 generated in the learning of step S200 in the teaching data database 320 ( S210 ).

[0074] In this way, by associating the visualization data 1000 related to a plurality of related websites with the source code and learning them, and storing them as the teaching data 3000 , it is possible to learn what kind of source code to output for what kind of visualization data.

[0075] After the storage of the teaching data 3000 in the teaching data database 320 is completed, the preliminary preparation for automatic generation of the web content is completed.

[0076] [Creation of web content using web content automatic generation system 1]

[0077] Next, creation of web content using the web content automatic generation system 1 according to the present embodiment will be described. Figure 6 1 is a flowchart showing a method for creating web contents using the web contents automatic generation system 1 .

[0078] The user who creates the web content first creates image data 4000 representing the web content he wants to create (step S310 ), and sends the created image data 4000 to the web content automatic generation system 1 (step S320 ).

[0079] The image data 4000 created in step S310 is first created as a wireframe by using document creation software or the like, or by arranging an image or text by handwriting, and then converted into image data.

[0080] When the image data 4000 is received in step S320 , the control unit 100 and the storage unit 300 of the web content automatic generation system 1 cooperate with each other to execute the image data registration module 118 to register the image data 4000 in the web content database 330 (step S330 ).

[0081] When the image data 4000 is registered in step S330 , the control unit 100 of the automatic content generation system 1 executes the image analysis module 119 to analyze the registered image (step S340 ).

[0082] When analyzing the image in step S340, the registered image data 4000 is scanned to collect information on the content and arrangement of the image and text. The scanning of the image data 4000 at this time is performed using the same method as the scanning of the imaged data 1000 in step S160.

[0083] Next, the control unit 100 and the storage unit 300 of the web page content automatic generation system 1 cooperate to execute the content source code generation module 120 to generate the content source code 5000 for generating the web page content (step S350). The content source code 5000 for generating the web page content includes data based on markup languages ​​such as HTML data and CSS data, and data related to images published in the web page content. In addition, JS data may also be included as needed.

[0084] When the content source code 5000 is generated in step S350, the information obtained by scanning the image data 4000 is compared with the plurality of teaching data 3000 stored in the teaching data database 320, and the source code of the latest teaching data 3000 is referenced to output the content source code 5000 for describing the content displayed by the image data 4000. In addition, post-processing such as image replacement and HTML syntax check is performed as needed to complete the generation of the content source code 5000.

[0085] When the content source code 5000 is generated in step S350, the control unit 100 of the web content automatic generation system 1 provides the generated content source code 5000 to the user terminal, and the automatic generation of the content is terminated (step S360).

[0086] [Modification 1]

[0087] In the web page content automatic generation system of this embodiment, in step S350, the content source code 5000 is generated based on the registered and analyzed image data 4000 and the latest teaching data 3000, but a plurality of content source codes 5000 may be generated based on a plurality of similar teaching data 3000 instead of the latest teaching data 3000. In this case, a plurality of different web page contents are generated, so that the best web page contents can be generated according to the user's preference.

[0088] By configuring as described above, it is possible to provide a web content automatic generation system that can easily create even a website having a complicated structure without editing a programming language.

[0089] The embodiments of the present invention are described above, but the present invention is not limited to the above embodiments. In addition, the effects described in the embodiments of the present invention are merely the best effects produced by the present invention, and the effects of the present invention are not limited to the effects described in the embodiments of the present invention.

[0090] [Second embodiment]

[0091] Next, a second embodiment will be described.

[0092] Figure 7 This is a block diagram for explaining the web content automatic generation system according to the second embodiment.

[0093] The web content automatic generation system 2 of the second embodiment includes a tag assigning unit 151 , a creating unit 153 , a learning unit 154 , a web design acquiring unit 155 , a content source code generating unit 156 , a communication unit 200 , a storage unit 300 , an input unit 400 , and a display unit 500 .

[0094] The tag assigning unit 151, the creating unit 153, the learning unit 154, the web page design obtaining unit 155, and the content source code generating unit 156 constitute one function of the control unit. Therefore, the tag assigning unit 151, the creating unit 153, the learning unit 154, the web page design obtaining unit 155, and the content source code generating unit 156 may also be constituted as a tag assigning module, a creating module, a learning module, a web page design obtaining module, and a content source code generating module constituting the control unit.

[0095] The labeling unit 151 acquires web page contents posted on a website open to the Internet, and labels the web page contents with labels indicating the contents for each content.

[0096] The label assigning unit 151 acquires a plurality of web page contents and assigns labels to the contents of each web page content. The label assigning unit 151 acquires a plurality of web page contents and assigns labels to the contents of each web page content. Figure 1 )) In the case where the image data obtained is stored in the storage unit 300, the image data can also be obtained from the storage unit 300 and labels can be assigned according to the contents of the image data.

[0097] Tags indicate classification of content such as images, texts, buttons, etc. For example, when web page content (image data) includes images, texts, and buttons, the tag assigning unit 151 assigns an image tag to the image, a text tag to the text, and a button tag to the button.

[0098] The label assigning unit 151 may assign labels to respective contents of the web page content based on, for example, an instruction from a user using the input unit 400 .

[0099] The production unit 153 produces a component that constitutes a part of the web page content based on the source code used to generate the web page content. The production unit 153 uses, for example, CSS code to produce the component. The component includes at least display information (images, text, colors, shapes, etc.) displayed in the web page content, and the user can indicate it. The component is, for example, a button. It should be noted that in addition to the above-mentioned buttons, the component may also be, for example, a background, a part of the background area, and their decorations. The production unit 153 produces multiple components with changed shapes, sizes, colors, etc. by changing CSS codes, etc.

[0100] The learning unit 154 creates teaching data based on the content labeled by the labeling unit 151. The learning unit 154 uses, for example, a convolutional neural network (CNN) to grasp the feature amount of the content and creates teaching data. That is, the learning unit 154 learns what the labeled content is, such as content (image) is an image, content (text) is text, and content (button) is a button.

[0101] In addition, the learning unit 154 produces teaching data based on the source code and the structure produced by the production unit 153. The learning unit 154 learns the CSS code produced by the production unit 153 using, for example, LSTM (Long Short-Term Memory) or the like. In addition, the learning unit 154 can also learn the structure through the production unit 153 using, for example, CNN or the like. The learning unit 154 produces teaching data by combining the learning results of CSS based on LSTM or the like with the learning results of the structure based on CNN or the like. As a specific example, the learning unit 154 produces teaching data by associating the CSS code with the structure.

[0102] The web design acquisition unit 155 acquires a web design for creating web content. The web design may be the image data described in the first embodiment. When the web design includes an image, the web design acquisition unit 155 acquires the image data (image data) and stores the image data in the storage unit 300 .

[0103] The content source code generating unit 156 acquires the content of the web page design based on the web page design acquired by the web page design acquiring unit 155 and the teaching data acquired by the learning unit 154 , and generates a source code corresponding to the content.

[0104] First, the content source code generation unit 156 analyzes the web page design based on the teaching data, thereby detecting the content in the web page design and assigning labels according to the content of the web page design. The labels can be the same as the labels assigned by the label assignment unit 151, such as images, texts, buttons, etc.

[0105] When there is content with a text tag in the web page design, the content source code generation unit 156 performs text detection on the content and generates text data. It should be noted that the content source code generation unit 156 may perform text detection based on the web page design acquired by the web page design acquisition unit 155 .

[0106] When part of the content of the web page design corresponds to the above-mentioned structure, the content source code generation unit 156 may generate a source code corresponding to the structure based on the teaching data. When there is a structure to which a button label is given as an example in the web page design, the content source code generation unit 156 generates a source code (for example, CSS code) for the structure based on the teaching data.

[0107] As described above, when the web page design includes an image, the web page design acquisition unit 155 may also acquire image data (image material) corresponding to the image. In this case, the content source code generation unit 156 may also generate position information for arranging an image based on the image data in the web page content based on the position of the image in the web page design. When there is content with an image tag in the web page design, the content source code generation unit 156 establishes a one-to-one correspondence between the image with the tag and the image data acquired by the web page design acquisition unit 155. That is, the content source code generation unit 156 matches the image with the image tag with the image data (image material) that matches the image.

[0108] It should be noted that the content source code generation unit 156 may also establish a one-to-one correspondence between the image and the image data (image material) included in the web page design acquired by the web page design acquisition unit 155. In this case, the higher the similarity between the image and the image data included in the web page design, the more likely the content source code generation unit 156 is to establish the correspondence.

[0109] The content source code generation unit 156 may also generate position information related to the position where the content in the web page content is configured based on the position of the content in the web page design, so as to correspond to the layout of the content in the web page design. The content source code generation unit 156 generates information (position information) related to the position where text, buttons, images, etc. are configured in the web page content based on the detection information when the content of the web page design is detected, as well as the results of text detection, button generation, and image matching. The position information is information used to configure text, buttons, and images (image data) in the web page content. The position information may also generate the coordinates of the coordinate system in the web page content based on the coordinates of the coordinate system in the web page design. In addition, the position information may be, for example, the coordinates of the center position of the content (text, button, image, etc.), or the coordinates of the upper left (or upper right, lower left, lower right) position.

[0110] That is, the content source code generation unit 156 generates position information for arranging text (text data) in the generated web page content so as to correspond to the position of the text in the web page design. In addition, the content source code generation unit 156 generates position information for arranging buttons in the generated web page content so as to correspond to the position of the buttons in the web page design. In addition, the content source code generation unit 156 generates position information for arranging images (image data) in the generated web page content so as to correspond to the position of the images in the web page design.

[0111] The content source code generation unit 156 may modify the label and position information based on the user's operation on the input unit 400. That is, the content source code generation unit 156 may modify the label information of text, image, button, etc., and their position information.

[0112] The content source code generation unit 156 generates the source code of the web page content by combining the CSS code of text, images (image data), buttons, etc. (components) and their position information, etc. The content source code generation unit 156 displays the web page content on the display unit 500 based on the generated source code.

[0113] The communication unit 200 , the storage unit 300 , the input unit 400 , and the display unit 500 of the second embodiment may have the same configuration as the communication unit 200 , the storage unit 300 , the input unit 400 , and the display unit 500 described in the first embodiment.

[0114] Next, a method for automatically generating web page contents according to a second embodiment will be described.

[0115] First, the processing flow when creating teaching data will be described.

[0116] Figure 8 This is a flowchart for explaining the flow of processing when creating teaching data according to the second embodiment.

[0117] In step ST301, the labeling unit 151 acquires web content posted on a website open to the Internet. The labeling unit 151 may automatically acquire a plurality of web content.

[0118] It should be noted that the web page content automatic generation system 2 may also replace the tag assigning unit 151 and obtain web page content as follows, similar to the first embodiment. That is, the website search unit (website search module 112: refer to Figure 1 ) accesses the Internet via the communication unit 200 and searches for websites. Web page display unit (web page display module 113: refer to Figure 1 ) Using a browser, the website found by the search is displayed on the display unit 500. The image data acquisition unit (image data acquisition module 114: see Figure 1 ) obtains the web page content of the website displayed by the display unit 500 as image data.

[0119] In step ST302, the label assigning unit 151 assigns labels to the web page content (image data) acquired in step ST301 according to each content. For example, when the web page content (image data) includes an image, the label assigning unit 151 assigns an image label to the image. For example, when the web page content (image data) includes text, the label assigning unit 151 assigns a text label to the text. For example, when the web page content (image data) includes a button, the label assigning unit 151 assigns a button label to the button.

[0120] In step ST303, the creation unit 153 creates a component constituting a part of the web page content based on source code such as CSS code. A component is, for example, a button, etc. The creation unit 153 creates a plurality of components with changed shapes, sizes, colors, etc.

[0121] In step ST304, the learning unit 154 creates teaching data based on the content labeled in step ST302. For example, the learning unit 154 creates teaching data based on the feature amount of the labeled content.

[0122] Furthermore, the learning unit 154 creates teaching data based on the source code used in step ST304 and the generated structure. For example, the learning unit 154 creates teaching data by associating the source code with the structure.

[0123] Next, the processing flow when automatically generating web page contents will be described.

[0124] Fig. 9 This is a flowchart for explaining the flow of processing when automatically generating web page contents according to the second embodiment.

[0125] In step ST401, the web design acquisition unit 155 acquires a web design for creating web content. In this case, if the web design includes an image, the web design acquisition unit 155 stores the image data (image data) in the storage unit 300 after the user uploads the image data.

[0126] In step ST402, the content source code generation unit 156 assigns tags to the contents of the web page design obtained in step ST401, and detects text for the contents to which the text tags indicating the text are assigned. That is, when there is content to which the text tags are assigned in the web page design obtained in step ST401, the content source code generation unit 156 performs text detection for the content using the learned model, and generates text data. It should be noted that the content source code generation unit 156 may also perform text detection based on the web page design obtained in step ST401.

[0127] In step ST403, when there is content corresponding to the component in the content of the web page design, the content source code generation unit 156 generates a source code corresponding to the component based on the teaching data generated in step ST305. That is, when there is a component to which a button label is given as an example in the web page content, the content source code generation unit 156 generates a source code (for example, a CSS code) of the component based on the teaching data.

[0128] In step ST404, when there is content with an image tag in the web page design, the content source code generation unit 156 establishes a one-to-one correspondence between the image with the tag and the image data acquired in step ST401. That is, the content source code generation unit 156 matches the image with the image tag with the image data that matches the image. It should be noted that the content source code generation unit 156 may also establish a one-to-one correspondence between the image and the image data included in the web page design acquired in step ST401.

[0129] In step ST405, the content source code generation unit 156 generates position information for arranging content in the web page content based on the position of the content described in the web page design acquired in step ST401. The content source code generation unit 156 generates information (position information) related to the position where text, components, and images are arranged in the web page content based on the detection information when the content of the web page design is detected, and the results of text detection, component (button) generation, and image matching.

[0130] In step ST406 , the content source code generation unit 156 may modify the position information and the like (label and position information) generated in step ST405 based on, for example, an operation of the input unit 400 by the user.

[0131] In step ST407, the content source code generation unit 156 generates source code corresponding to the web page design obtained in step ST401 based on the text detected in step ST402, the source code of the structure generated in step ST403, the image (image data) matched in step ST404, the position information generated in step ST405, and the correction performed in step ST406.

[0132] Next, the effects of the second embodiment will be described.

[0133] The web page content automatic generation system 2 comprises: a label assigning unit 151, which obtains web page content and assigns labels to the web page content according to each content; a production unit 153, which produces a structure that constitutes a part of the web page content based on the source code used to generate the web page content; a learning unit 154, which obtains teaching data based on the content labeled by the label assigning unit 151 and based on the source code and the structure produced by the production unit 153; a web page design acquisition unit 155, which obtains a web page design; and a content source code generation unit 156, which obtains the content of the web page design based on the web page design obtained by the web page design acquisition unit 155 and the teaching data obtained by the learning unit 154, and generates source code corresponding to the content.

[0134] Thus, according to the web content automatic generation system 2, by learning the teaching data in advance, the source code can be automatically generated when the web page design is input. Therefore, the web content automatic generation system 2 can generate the web content desired by the user.

[0135] In the web content automatic generation system 2 , when a part of the content of the web page design corresponds to a structure, the content source code generation unit 156 may generate a source code corresponding to the structure based on the teaching data.

[0136] Thus, even when the web content desired by the user includes a component such as a button, the automatic content generation system 2 can generate web content including the component.

[0137] In the web content automatic generation system 2, the content source code generation unit 156 may generate position information related to the position where the content in the web content is arranged based on the position of the content in the web design so as to correspond to the layout of the content in the web design.

[0138] Thus, the web content automatic generation system 2 can generate web content according to the web page design.

[0139] In the web page content automatic generation system 2, when the web page design includes an image, the web page design acquisition unit 155 may also acquire image data corresponding to the image. In this case, the content source code generation unit 156 may also generate position information for arranging the image based on the image data in the web page content based on the position of the image in the web page design.

[0140] Thus, even when the web page design includes images, the web page content automatic generation system 2 can generate web page content including images corresponding to the web page design.

[0141] Each unit of the web page content automatic generation system 2 described above may also be implemented as a function of a computer processing device or the like. That is, the tagging unit 151, the creation unit 153, the learning unit 154, the web page design acquisition unit 155, and the content source code generation unit 156 of the web page content automatic generation system 2 may also be implemented as a tagging function, a creation function, a learning function, a web page design acquisition function, and a content source code generation function based on a computer processing device or the like, respectively.

[0142] The web page content automatic generation program can enable the computer to realize the above functions. The web page content automatic generation program can also be recorded in a non-transitory recording medium such as an external memory or an optical disk that can be read by a computer.

[0143] In addition, as described above, each part of the web page content automatic generation system 2 can also be implemented by a computer processing device, etc. The processing device, etc. is composed of, for example, an integrated circuit, etc. Therefore, each part of the web page content automatic generation system 2 can also be implemented as a circuit constituting a processing device, etc. That is, Fig.10 As shown, the label assigning unit 151, the production unit 153, the learning unit 154, the web page design acquisition unit 155 and the content source code generation unit 156 of the web page content automatic generation system 2 can also be implemented as a label assigning circuit 151a, a production circuit 153a, a learning circuit 154a, a web page design acquisition circuit 155a and a content source code generation circuit 156a constituting a computer's processing device, etc.

[0144] In addition, the communication unit 200, the storage unit 300, the input unit 400, and the display unit 500 of the web page content automatic generation system 2 may be configured by, for example, an integrated circuit or the like, as the communication circuit 200a, the storage circuit 300a, the input circuit 400a, and the display circuit 500a (see Fig.10 ) In addition, the communication unit 200, storage unit 300, input unit 400 and display unit 500 of the web page content automatic generation system 2 may also be composed of a plurality of devices, such as a communication device, a storage device, an input device and a display device.

[0145] Description of reference numerals:

[0146] 1.2 Web Content Automatic Generation System

[0147] 100 Control Department

[0148] 112 Website Search Module

[0149] 114 Image data acquisition module

[0150] 115 Source code acquisition module

[0151] 116 Learning Modules

[0152] 119 Image parsing module

[0153] 120 Content source code generation module

[0154] 151 Labeling Department

[0155] 153 Production Department

[0156] 154 Learning Department

[0157] 155 Web Design Acquisition Department

[0158] 156 Content Source Code Generation Department

[0159] 200 Ministry of Communications

[0160] 300 Storage

[0161] 310 Web Database

[0162] 330 Web Content Database

Claims

1. A web page content automatic generation system, which automatically generates web page content described in HTML, wherein: The webpage content automatic generation system has: an image data acquisition unit that scans and connects screens displayed on a plurality of websites that are open and related to the Internet in an end-to-end manner based on the search information, thereby collecting the screens as teaching data, and acquires the web page contents published in the collected websites as one image data; A source code acquisition unit that acquires source code including HTML data, CSS data, and JS data constituting the content of the web page; a learning unit that learns the imaged data acquired by the imaged data acquisition unit and the source code corresponding to the imaged data acquired by the source code acquisition unit, and learns which source code is output for which imaged data, thereby generating a learned model; An image data registering unit that inputs the content of a web page to be created as image data; as well as The content source code generation unit generates a content source code corresponding to the input image data based on the input image data and the learned model generated by the learning unit.

2. The webpage content automatic generation system according to claim 1, wherein: The web page content automatic generation system further includes a website search unit for searching the website. The website search unit automatically searches for websites published on the Internet using a search engine based on input search information.

3. A web page content automatic generation system, wherein: The webpage content automatic generation system has: a labeling unit that acquires web page contents posted on a website open to the Internet and labels the web page contents for each content including images, texts, and buttons; A production unit that produces components including display information, buttons, background, a part of the background area, and decorations that constitute a part of the web page content based on the source code for generating the web page content, and changes information including shape, size, and color of the components; a learning unit that acquires teaching data based on the content labeled by the labeling unit and based on the source code and the structure produced by the producing unit; A web page design acquisition unit that acquires web page designs for producing web page contents; as well as A content source code generating unit assigns different labels according to different contents of the web page design, classifies the different labels, and acquires the contents of the web page design based on the web page design acquired by the web page design acquiring unit and the teaching data acquired by the learning unit, and generates source code corresponding to the contents.

4. The webpage content automatic generation system according to claim 3, wherein: The content source code generating unit generates a source code corresponding to the structure based on teaching data when a part of the content of the web page design corresponds to the structure.

5. The webpage content automatic generation system according to claim 3 or 4, wherein: The content source code generation section generates position information related to a position where the content in the web page content is arranged based on the position of the content in the web page design so as to correspond to the layout of the content of the web page design.

6. The webpage content automatic generation system according to claim 5, wherein: The web page design acquisition unit acquires image data corresponding to the image when the web page design includes an image. The content source code generation unit generates position information for arranging an image based on the image data in the web page content based on the position of the image in the web page design.

Images