Content generation device, content generation method, and computer program

Abstract

Increase the amount of content on social media and revitalize it. [Solution] A content generation device 1 that generates generated content to be published on an SNS, comprising a processor, in which, in a post acceptance step, the processor displays an input screen that accepts input of posted content to be posted on the SNS, in a content setting step, the processor accepts specification of generation conditions for the generated content, including at least the data form of the generated content, in a generation instruction step, the processor issues a generation instruction to an artificial intelligence to generate generated content in accordance with the generation conditions based on the posted content, and in a display step, the generated content output by the artificial intelligence is displayed, and the artificial intelligence has a trained model that takes at least the posted content as input and generates the generated content as output.

Description

[Technical Field]

[0001] The present invention relates to a technology for generating content on a social networking service (SNS). [Background technology]

[0002] In recent years, many users have joined SNS, and various information is being transmitted within SNS. SNS becomes more active when a large amount of useful content is posted.

[0003] Patent Document 1 discloses a technology for creating article content for blogs and the like that takes search engine optimization (SEO) into consideration based on information such as words received from users. Patent Document 2 discloses a system that acquires related keywords that contribute to SEO based on information entered by users and preset reference information, and generates an article structure for content based on the related keywords. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2022-181236 [Patent Document 2] Patent No. 7441576

[0005] All of the above patent documents have room for improvement in terms of revitalizing SNS. Summary of the Invention [Problem to be solved by the invention]

[0006] An object of the present invention is to provide a content generation device that can further activate an SNS by increasing the amount of content on the SNS. [Means for solving the problem]

[0007] In order to achieve the above-mentioned object, a content generation device according to one aspect of the present invention is a content generation device that generates generated content to be published on a social networking site, and includes a processor, wherein the processor, in a post acceptance step, displays an input screen that accepts input of posted content to be posted on the social networking site; in a content setting step, accepts specification of generation conditions for the generated content, including at least the data form of the generated content; in a generation instruction step, issues a generation instruction to an artificial intelligence to generate the generated content in accordance with the generation conditions based on the posted content; and in a display step, displays the generated content output by the artificial intelligence, and the artificial intelligence has a trained model that takes at least the posted content as input and outputs the generated content.

[0008] A selection step may further be performed to select the posted content that includes a predetermined designated word, and in the generation instruction step, a generation instruction may be given to the artificial intelligence to generate the generated content that includes the designated word based on the posted content selected in the selection step.

[0009] A storage step may further be performed to store the designated word and co-occurring words associated with the designated word, and in the generation instruction step, a generation instruction may be given to the artificial intelligence to generate the generated content including the designated word and the co-occurring words by referring to the co-occurring words associated with the designated word.

[0010] The generation instruction step may include issuing a generation instruction to the artificial intelligence to generate the generated content based on the posted content.

[0011] In order to achieve the above-mentioned object, a content generation method according to another aspect of the present invention is a content generation method for generating generated content to be published on a social networking site, which includes a post acceptance step of displaying an input screen for accepting input of posted content to be posted on the social networking site by a computer, a content setting step of accepting specification of generation conditions for the generated content, including at least the data form of the generated content, a generation instruction step of issuing a generation instruction to an artificial intelligence to generate the generated content in accordance with the generation conditions based on the posted content, and a display step of displaying the generated content output by the artificial intelligence, wherein the artificial intelligence has a trained model that receives at least the posted content as input and the generated content as output.

[0012] In order to achieve the above-mentioned object, a computer program according to yet another aspect of the present invention is a computer program for generating generated content to be published on a social networking site, and causes a computer to execute a post acceptance step of displaying an input screen for accepting input of posted content to be posted on the social networking site; a content setting step of accepting specification of generation conditions for the generated content, including at least the data form of the generated content; a generation instruction step of issuing a generation instruction to an artificial intelligence to generate the generated content in accordance with the generation conditions based on the posted content; and a display step of displaying the generated content output by the artificial intelligence, wherein the artificial intelligence has a trained model that receives at least the posted content as input and the generated content as output.

[0013] The computer program can be provided by downloading it via a network such as the Internet, or by recording it on various computer-readable recording media. [Effects of the Invention]

[0014] According to the present invention, the amount of content on an SNS can be increased, thereby further activating the SNS. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a functional block diagram illustrating an embodiment of a content generation device according to an embodiment of the present invention. [Figure 2] 10 is a flowchart showing a first example of a processing flow of the content generator. [Figure 3] 10 is a flowchart showing a second example of the processing flow of the content generator. [Figure 4] 3A and 3B are diagrams showing a first example and a second example of a screen displayed on a user terminal by the content generation device. [Figure 5] 10A and 10B are diagrams showing a third example and a fourth example of a screen displayed on the user terminal by the content generation device. [Figure 6] FIG. 4 is a diagram showing a first example of a screen displayed on an administrator terminal by the content generator. [Figure 7] FIG. 10 is a diagram showing a second example of a screen displayed on the administrator terminal by the content generator. [Figure 8] FIG. 10 is a diagram showing a third example of a screen displayed on the administrator terminal by the content generator. [Figure 9] FIG. 10 is a diagram showing a fourth example of a screen displayed on the administrator terminal by the content generator. [Figure 10] FIG. 10 is a diagram showing a fifth example of a screen displayed on the administrator terminal by the content generator. [Figure 11] FIG. 10 is a diagram showing a sixth example of a screen displayed on the operator terminal by the content generator. DETAILED DESCRIPTION OF THE INVENTION

[0016] An embodiment of the present invention will be described with reference to FIG. The content generation device 1 shown in FIG. 1 is a device that manages an SNS and generates content to be posted on the SNS (hereinafter, the SNS managed by the content generation device 1 will also be referred to as the "managed SNS"). In this SNS, multiple users register accounts on the SNS and log in to their accounts to view and post content via their respective devices. Individual users participating in the managed SNS can post content within the managed SNS, and other users participating in the managed SNS can view this content. The content is composed of, for example, any one or a combination of text data, image data, video data, and music data, and is data that includes information such as product and service introductions or various opinions.

[0017] The content generation device 1 is, for example, a server, and is connected to an operator terminal 20 and a user terminal 30 via a network NW. Although only one operator terminal 20 and one user terminal 30 are shown in the figure, multiple operator terminals 20 and multiple user terminals 30 are connected and can participate in the SNS. The operator is, for example, the operator of a community set within the management SNS. In the following explanation, it is assumed that the management SNS has multiple communities, and that an operator set for each community has the authority to set the conditions for generating content generated from posted content posted within the community and to decide whether or not to make the generated content public. Note that this embodiment is merely an example, and an operator may be set who has the authority to reference posted content and generate content regardless of the community to which it is posted. The operator terminal 20 and the user terminal 30 may have the same hardware configuration.

[0018] The operator terminal 20 and the user terminal 30 each include a display unit 21, 31 that displays information received from the content generator 1, and an input unit 22, 32 that accepts input from the operator or user, respectively. The display units 21, 31 are configured, for example, by a display, and the input units 22, 32 are configured, for example, by a touch panel or a keyboard. The user terminal may be a personal computer or a mobile terminal such as a tablet terminal or a smartphone.

[0019] In this embodiment, it is assumed that there may be a plurality of client terminals equipped with the content generation device 1. Also, there may be a plurality of terminals equipped with the content generation device 1. Furthermore, there may be a plurality of servers equipped with the content generation device 1. Also, the functions of the content generation device 1 may be realized as a system by a plurality of configurations connected via a network. Some or all of the functions of the content generation device 1 may be realized by cloud computing.

[0020] ●Function configuration The content generation device 1 is composed of a storage medium such as a memory, a processor, a communication module, an input / output interface, etc., and is configured to realize the functional blocks shown in Fig. 1 by the processor executing a computer program recorded on the storage medium. The storage medium is a computer-readable recording medium and may include storage devices such as RAM (random access memory), ROM (read only memory), a disk drive, SSD (solid state drive), and flash memory. Here, non-transitory storage devices such as ROM, disk drive, SSD, and flash memory may be included in the content generation device 1 as storage devices separate from the memory.

[0021] The content generation device 1, with the above-described hardware configuration, mainly comprises a display control unit 11, a content extraction unit 12, a generation condition setting unit 13, a posting reception unit 14, an AI control unit 15, a data management unit 16, a memory unit 18, and a communication processing unit 19.

[0022] The display control unit 11 is a functional unit that displays posted content sent by users via the user terminal 30 on other user terminals 30. The display control unit 11 also displays a screen on the management terminal 20 that accepts setting of conditions for content to be generated. Furthermore, the display control unit 11 displays generated content generated by the content generation device 1 on the management terminal 20 and the user terminal 30.

[0023] The content extraction unit 12 is a functional unit that extracts posted content that is the source of content generated by the content generation device 1. For example, the content extraction unit 12 extracts posted content in accordance with predetermined conditions for extracting posted content. For example, the content extraction unit 12 considers content for which it has been agreed that the copyright will be waived as the source of content, and does not consider posted content for which the copyright has not been waived as the source of content.

[0024] The content extraction unit 12 may, for example, extract posted content that includes a predetermined designated word and determine that posted content as the source of generation. The designated word may be, for example, a word related to an issue for which the administrator wishes to increase access to the web page, i.e., a so-called SEO word. By generating content based on posted content that includes SEO words, it is possible to increase the amount of content with the desired content and realize effective SEO measures.

[0025] Furthermore, the content extraction unit 12 may use all posted content for which the copyright has been waived as the source of the generated content without referring to the specified words, or may use posted content extracted under appropriate conditions as the source of the generated content. The content extraction unit 12 may also use posted content with a predetermined number of reactions or comments or more as the source of the generated content.

[0026] Posted content is content posted by users of the managed SNS, known as user-generated content (UGC). A configuration for generating content based on user-generated content allows for the generation of original content that meets user needs. Furthermore, user-generated content tends to attract more interest from other users than content created by companies. This configuration allows for the easy generation of content that attracts the interest of other users.

[0027] The generation condition setting unit 13 is a functional unit that sets the generation conditions for the generated content. The generation condition setting unit 13 sets the generation conditions, for example, in response to input from the administrator via the management terminal 20. For example, in a mode in which posted content to be generated is extracted using a specified word, the generation condition setting unit 13 may determine co-occurring words to be included in the generated content in response to the specified word. Co-occurring words are, for example, words that are highly relevant to the specified word and are likely to attract interest in combination with the specified word. For example, if the specified word is "Ueno Park," co-occurring words are words such as "opening hours," "admission fee," and "access." Multiple co-occurring words may be set for one specified word. A configuration in which the generated content includes both the specified word and the co-occurring words not only ensures the quality of the information in the generated content, but also generates content that is more appealing to users who are interested in the specified word.

[0028] The content generation conditions may include the expected article title, the target readership of the generated content (also known as the "target"), the purpose of reading the article, and the volume of the generated content. The volume of the generated content may include, for example, the number of paragraphs and the number of characters. Furthermore, the output format of the content may include, for example, information indicating that the content is in an HTML structure.

[0029] The content generation conditions may also include the number of specified words or the number and type of co-occurring words included in one generated content. Note that an instruction to output the number of specified words or the number and type of co-occurring words useful for SEO measures may be generated for an appropriate artificial intelligence unit having a trained model that has learned the number of specified words or the number and type of co-occurring words and the SEO ranking, and the output result from the artificial intelligence unit may be included in the content generation conditions.

[0030] Furthermore, the generation condition setting unit 13 may determine the number of generated contents to be generated. The number of generated contents may be the number of contents generated from one posted content, or may specify the number to be generated from multiple posted contents. The generation condition setting unit 13 may also determine the data format of the generated content. The data format may be text data, image data, video data, music data, or a combination of these.

[0031] The posting acceptance unit 14 is a functional unit that accepts posted content from appropriate users via the user terminal 30. The data format of the posted content may be text data, image data, video data, music data, or a combination of these. The posting acceptance unit 14 may accept multiple posted content from one user. When the posting acceptance unit 14 accepts the posted content, a message is displayed on the user terminal 30 that sent the posted content, asking the user whether to waive the copyright of the posted content. When the posting acceptance unit 14 accepts an input from the user terminal 30 agreeing to the copyright waiver, the posting acceptance unit 14 stores information indicating that the copyright has been waived in the memory unit 18 in association with the posted content.

[0032] The AI ​​control unit 15 is a functional unit that controls input to the artificial intelligence unit 200. Various instructions generated by the AI ​​control unit 15 are, for example, prompts, but are not limited to these as long as they are instructions that the artificial intelligence unit 200 can interpret.

[0033] The AI ​​control unit 15 generates a content generation instruction for the artificial intelligence unit 200 to generate content by, for example, referring to posted content that is the source of the generated content. There may be one or more posted contents that are the source of one generated content.

[0034] The AI ​​control unit 15 generates a generation instruction to generate content according to the generation conditions set by the content extraction unit 12. The AI ​​control unit 15 may also generate a generation instruction to generate content related to a specified word. The AI ​​control unit 15 may include a specification of a writing style in the generation instruction, and may generate a generation instruction to generate text content in a style similar to that of news media, for example. Furthermore, the AI ​​control unit 15 may generate an instruction to generate appropriate information, such as meta keyword tags or meta description tags, that is added to the generated content and is not visible on the content display page. Furthermore, by including an instruction to add descriptions of products and services introduced in the user-generated content and include a certain degree of relative information, it is possible to generate content that captures user needs while ensuring the quality of the information.

[0035] The artificial intelligence unit 200 is an artificial intelligence (AI) equipped with a language model such as a transformer, including a bidirectional and auto-regressive transformer (BART), a bidirectional encoder representations from transformers (BERT), or a generative pretrained transformer (GPT-1, GPT-2, GPT-3, GPT-4, and GPT-4o), particularly a learning model (also referred to as a machine learning model) such as a large-scale language model (LLM). The learning model is a learning model based on a machine learning algorithm. Specific examples of machine learning algorithms include nearest neighbor algorithms, naive Bayes algorithms, decision trees, and support vector machines. Another example is deep learning, which uses a neural network to generate features and connection weighting coefficients for learning. The artificial intelligence unit 200 may be equipped with an appropriate image generation model, video generation model, or music generation model. The artificial intelligence unit 200 can apply the above algorithms as appropriate.

[0036] For example, the artificial intelligence unit 200 refers to an appropriate database DB based on a generation instruction generated by the AI ​​control unit 15, inputs the posted content, and outputs generated content related to the posted content. For example, the artificial intelligence unit 200 generates one or more generated content items for each posted content item. With this configuration, various pieces of content related to the posted content can be easily generated. By publishing this content item on a managed SNS or the like, the number of pieces of content on the SNS can be increased, and the SNS can be activated.

[0037] The artificial intelligence unit 200 may input the posted content, the specified word, and the co-occurring words based on a generation instruction generated by the AI ​​control unit 15, and output generated content related to the posted content. The artificial intelligence unit 200 outputs generated content for each co-occurring word. With this configuration, it is possible to generate content from various perspectives according to the co-occurring words in a balanced manner.

[0038] The generation instruction may also include the data format of the content to be generated, and the artificial intelligence unit 200 generates content in the data format instructed by the AI ​​control unit 15. In this case, the artificial intelligence unit 200 may generate image data or video data based on posted content of text data, or may generate posted content of text data or video data based on image data. The generated image data may be an image to be displayed together with the text data, or may be image data of a predetermined size that can also be used for product advertisement banners.

[0039] The trained model not only obtains appropriate training data in advance, but also can undergo additional training as appropriate. The training data may be provided by an administrator or the like, or may include information collected from the Internet or the like. The trained model may be trained using posted content received by the posting receiving unit 14, for example. The trained model may be trained using posted content for which the copyright has been waived, and may not be trained on posted content for which the copyright waiver has not been agreed to. The artificial intelligence unit 200 may have a trained model for each community in the managed SNS that has trained on the posted content within that community. In this case, the AI ​​control unit 15 may generate an instruction to generate content using the trained model trained on the posted content posted to the target community of the generated content.

[0040] The generated content is associated with the identification information of the administrator of the community to which the posted content is posted and stored in the storage unit 18. In addition, the generated content including the designated word is further associated with the designated word and stored in the storage unit 18.

[0041] The data management unit 16 is a functional unit that manages the generated content obtained from the artificial intelligence unit 200. The data management unit 16 displays the generated content on the operator terminal 20 via the display control unit 11 and accepts instructions regarding whether to make the content public and the scope of public disclosure. The operator terminal 20 can input instructions regarding whether to make the content public and the scope of public disclosure for each piece of generated content. The data management unit 16 may also set the disclosure mode of the generated content within the managed SNS, and may set the generated content to be displayed as an advertisement, or may display the generated content in the same manner as a user post. When displaying the generated content as an advertisement, the data management unit 16 may attach a predetermined link that is previously stored and associated with the product or service that is the subject of the advertisement to the advertisement.

[0042] The data management unit 16 publishes the generated content selected for publication according to the selected publication range. For example, if the publication range is specified within the administrative SNS, the data management unit 16 issues a URL within the administrative SNS and attaches a link to a specified webpage. The specified webpage may be, for example, the homepage of the community to which the original posted content was posted, or an appropriate page within the community. Furthermore, if an external SNS different from the administrative SNS is specified as the publication range, the data management unit 16 posts the generated content to the external SNS via a posting account pre-registered on the external SNS. In this case, the generated content may include a tag or link indicating the administrative SNS. This configuration can appeal to viewers who do not use the administrative SNS and is expected to attract new users to the administrative SNS.

[0043] The data management unit 16 may extract external SNSs to which the generated content can be posted according to the data type of the generated content and display them selectably on the management terminal 20. External SNSs may have restrictions on the data type that can be posted, such as SNSs that only allow videos to be posted. In this regard, a configuration that can generate content of various data types according to a generation instruction makes it possible to post to a variety of external SNSs. This in turn increases the number of new users who flow into the managed SNS from external SNSs, further activating the managed SNS. Furthermore, a configuration that extracts and displays external SNSs to which the generated content can be posted according to the data type makes it easier for the administrator to post externally.

[0044] In this embodiment, the administrator selects whether or not to allow the disclosure of generated content via the management terminal 20, but the administrator may also be configured to make all generated content public. Also, the scope of disclosure of generated content may be set in advance.

[0045] According to the above-described configuration, a variety of content can be generated based on content posted to the managed SNS. According to a configuration based on posted content, more realistic and highly original content can be generated. Furthermore, according to a configuration in which content is generated based on posted content, it is possible to generate highly sensitive content that is in line with popular content and phrases. Furthermore, according to a configuration in which content including designated words is generated based on posted content that includes designated words, it is possible to generate natural content that is relevant to the designated words. Ultimately, an increase in high-quality content can increase the community value of the SNS and revitalize the SNS. Furthermore, according to a configuration in which generated content is displayed as an advertisement, a variety of advertising copy can be easily generated. Ultimately, it is possible to increase sales through advertising on the SNS.

[0046] The storage unit 18 is a database that stores information related to the management of the SNS. For example, the storage unit 18 stores generated content, the generation conditions of the generated content, designated words, and identification information of the administrator who set the designated words, in association with each other. The storage unit 18 may also store identification information of the posted content that was the source of the generated content, in association with the generated content. Furthermore, the storage unit 18 may store information such as whether the generated content is public and the extent of its public availability.

[0047] The communication processing unit 19 is a functional unit that transmits and receives information to and from the operator terminal 20 and the user terminal 30 via the network NW, and for example, receives login information and content posts from the user terminal 30. The communication processing unit 19 also transmits content generation instructions to the artificial intelligence unit 200 and receives generated content from the artificial intelligence unit 200. Furthermore, the communication processing unit 19 transmits pages within the SNS, content submitted by users, and generated content generated by the content generation device 1 to the user terminal 30.

[0048] ●Processing flow (1) A first example of a series of processes for generating content in the content generation device 1 according to this embodiment will be described with reference to Fig. 2. The first example is an example of a process for generating content without referring to a designated word. Before starting this process, the content generation device 1 registers the user's SNS information. More specifically, the content generation device 1 prompts the user to enter a user ID and password on an appropriate user registration page of the management SNS, and then registers an SNS account linked to the user ID and password.

[0049] First, when content is posted from the user terminal 30 (step S11), the content generator 1 displays a message on the user terminal 30 asking whether to waive copyright (step S12). When consent to waive copyright is input from the user terminal 30 (step S13), the process proceeds to step S14. If consent to waive copyright is not input, the posted content is made public on a predetermined page of the management SNS, and the process ends.

[0050] In step S14, a generation instruction to generate content is transmitted to the artificial intelligence unit 200. Next, the content generation device 1 acquires the generated content from the artificial intelligence unit 200. The content generation device 1 transmits the generated content to the operator terminal 20 (step S15).

[0051] The operator terminal 20 displays a screen for accepting input of whether or not to make each generated content public. If public is selected through the operator terminal 20 (Y in step S16), the selection of the public range is accepted (step S17), and the process proceeds to step S18. If not public is selected through the operator terminal 20 (N in step S16), the process proceeds to step S18 without going through step S17.

[0052] In step S18, the content generator 1 performs different processing depending on whether or not to make the content public, as selected by the management terminal 20. That is, if public disclosure is selected (Y in step S18), the generated content is made public according to the selected range of disclosure (step S19). If no public disclosure is selected (N in step S18), the generated content is stored in the storage unit 18 (step S20), and the processing ends. Note that step S20 may be performed after step S14.

[0053] ●Processing flow (2) A second example of a series of processes for generating content in the content generation device 1 according to this embodiment will be described with reference to Figure 3. The second example is an example of a process for generating content from posted content that includes a specified word that has been designated in advance. Note that a description of processes similar to those in the first example will be omitted as appropriate.

[0054] In the second example, when consent to the waiver of copyright is input from the user terminal 30 in step S13, the process proceeds to step S31, where the content generation device 1 determines whether the posted content contains the designated word (step S31). If the posted content contains the designated word (Y in step S32), the process proceeds to step S33. If the posted content does not contain the designated word (Y in step S32), the process ends without generating content. In step S33, an instruction to generate content containing the extracted designated word and the co-occurring words stored in association with the designated word is transmitted to the artificial intelligence unit 200. The process then proceeds to step S16.

[0055] If the posted content does not contain the designated word in step S32, the process may proceed to step S14 shown in Fig. 2, where content that does not contain the designated word may be generated. In this case, generated content that contains the designated word and generated content that does not contain the designated word are displayed separately on the operator terminal 20.

[0056] ●Screen example Here, examples of screen transitions in the operator terminal 20 and the user terminal 30 will be described with reference to Figures 4 to 11. In the description, operations for selecting buttons or icons on the screen indicate various user operations, such as tapping or clicking, that are adopted depending on the functions of the terminal.

[0057] 4(a) is an example of a screen G10 displayed on the user terminal 30 on which a user posts posted content. Screen G10 displays input areas for the title and text of the posted content. Screen G10 also displays the category to which the posted content will be posted. Screen G10 also displays an icon for starting the attachment of a data file to be posted separately from the text data. When the user has completed inputting the content to be posted, he or she selects the post button displayed on screen G10. When the post button is selected, the screen transitions to screen G20 shown in FIG. 4(b).

[0058] Screen G20 shown in Figure 4(b) is a screen that asks the user to agree to the waiver of copyright for the posted content. The screen displays an explanation about the secondary use of the posted content, as well as buttons to select whether to post with or without consent, or to return to editing the posted content.

[0059] 5 is an example of a registration screen G30 displayed on the administrator terminal 20 and accepting input for registering a designated word. The registration screen G30 displays input areas for the newly registered designated word, co-occurring words associated with the designated word, expected article title, article target, purpose of reading the article, and article length. One or more co-occurring words, article target, and purpose of reading the article may be input. The article length may be specified by the number of characters, or may also accept the selection of an estimated reading time.

[0060] FIG. 6 is an example of a screen G40 that displays a list of generated content. Screen G40 displays information about the data format of the generated content and the nature of the content, such as whether the generated content is a product advertisement. Screen G40 also displays an image, a title, and at least a portion of the text for each piece of generated content. Furthermore, each piece of generated content is displayed so that it can be selected whether to make it public. Furthermore, screen G40 displays information about the posted content that generated the listed generated content, such as the title of the posted content, the posting date and time, the user name, the number of reactions, and the number of comments.

[0061] FIG. 7 is an example of a screen G50 that is displayed when a button to display details of generated content is selected on screen G40 in FIG. 6. Screen G50 displays the title, article length, main text, images, etc. of the generated content. Also displayed is information about the posted content that was the source of the generated content. If there is more than one posted content that was the source of the generated content, information about the multiple posted contents may be displayed. Also displayed on screen G50 is an edit button that starts editing the generated content. With this configuration, by allowing the operator to edit the generated content automatically generated by artificial intelligence, content that meets the operator's intentions can be easily created.

[0062] Figure 8 is another example of a screen displaying a list of generated content. Screen G60 shown in Figure 8 displays the associated and stored co-occurring words and the number of generated contents for each specified word. Note that the specified words are referred to as SEO words in the figure. Also, an article list button that displays a list of generated content is displayed for each specified word.

[0063] FIG. 9 shows screen G70, which is displayed when the article list button for the specified word "Okinawa Trip" is selected on screen G60. Screen G70 displays, for each generated content, the title, co-occurring words included in the instructions for generation, whether or not the content is published, and its SEO ranking. The SEO ranking may also be displayed for each search engine. This configuration clarifies the SEO value associated with co-occurring words and can help operators evaluate co-occurring words.

[0064] 10 shows an example of a screen G80 that is displayed when the user selects the button for displaying details of the generated content on the screen G70. The screen G80 displays the title, length, text, images, etc. of the generated content. It also displays information about the posted content that generated the generated content.

[0065] 11 is a screen G90 showing another example of displaying details of generated content. Screen G90 displays appropriate information to be added to the generated content, such as meta keyword tags and meta description tags.

[0066] Effects of this embodiment According to this embodiment, the amount of content on the SNS can be increased, and the SNS can be further activated. [Explanation of symbols]

[0067] 1 Content generation device 11 Display control unit 12 Content Extraction Unit 13 Generation condition setting section 14 Submission Reception Department 15 AI control section 16 Data Management Department 18 Memory section 19 Communication processing unit 20 Operator terminal 30 User terminals

Claims

1. A content generation device that generates generated content to be published on an SNS, a processor, the processor comprising: In the post receiving step, an input screen for receiving input of post content to be posted to the SNS is displayed, In the content setting step, a specification of a generation condition of the generated content including at least a data form of the generated content is accepted; In the generation instruction step, a generation instruction is given to an artificial intelligence to generate the generated content in accordance with the generation conditions based on the posted content; In the display step, the generated content output by the artificial intelligence is displayed; The artificial intelligence has a trained model that uses at least the posted content as input and the generated content as output. Content generation device.

2. a selection step of selecting the posted content including a predetermined designated word; In the generation instruction step, a generation instruction is given to the artificial intelligence to generate the generated content including the designated word based on the posted content selected in the selection step. The content generation device according to claim 1.

3. further performing a storage step of storing the designated word and a co-occurring word linked to the designated word; In the generation instruction step, a generation instruction is given to the artificial intelligence to generate the generated content including the designated word and the co-occurring word by referring to co-occurring words linked to the designated word.

3. The content generating device according to claim 2.

4. In the generation instruction step, a generation instruction is given to the artificial intelligence to generate the generated content based on the posted content. The content generation device according to claim 1.

5. A content generation method for generating generated content to be published on an SNS, comprising: By computer, a post receiving step of displaying an input screen for receiving input of post content to be posted to the SNS; a content setting step of receiving a designation of a creation condition of the created content, the designation including at least a data form of the created content; a generation instruction step of issuing an instruction to an artificial intelligence to generate the generated content in accordance with the generation conditions based on the posted content; a display step of displaying the generated content output by the artificial intelligence; Run The artificial intelligence has a trained model that uses at least the posted content as input and the generated content as output. Content generation methods.

6. A computer program for generating generated content to be published on an SNS, On the computer, a post receiving step of displaying an input screen for receiving input of post content to be posted to the SNS; a content setting step of receiving a designation of a creation condition of the created content, the designation including at least a data form of the created content; a generation instruction step of issuing an instruction to an artificial intelligence to generate the generated content in accordance with the generation conditions based on the posted content; a display step of displaying the generated content output by the artificial intelligence; Execute The artificial intelligence has a trained model that uses at least the posted content as input and the generated content as output. Computer program.

Images