SNS posting system

Abstract

The goal is to enable the generation of a second post, which will be posted simultaneously with the first post, through a simple operation. [Solution] The SNS posting system generates a first post content to be uploaded to a first SNS account based on the user's input, generates a second post content to be uploaded to a second SNS account by changing the input results according to the user's settings or operations, and uploads the first post content to the first SNS account and the second post content to the second SNS account according to the user's operations.

Description

Technical Field

[0001] The present invention relates to an SNS posting system, and more particularly to an SNS posting system that enables simultaneous posting to multiple SNSs.

Background Art

[0002] There is a known system that enables simultaneous posting of the same content to multiple SNSs (Social Networking Services) with a single operation. Patent Document 1 discloses an example of such a system.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the above conventional system, the content that can be simultaneously posted with a single operation is always the same. It does not support cases where it is desired to use Japanese and English posts depending on the SNS, or cases where the content of the post is slightly changed according to the character limit of the SNS, which is inconvenient.

[0005] Therefore, one object of the present invention is to provide an SNS posting system that can generate a second posting content to be simultaneously posted with a first posting content by a simple operation with a single operation.

[0006] By the way, in order to enable simultaneous posting of content with the same meaning but different description languages with a single operation, it is conceivable to use automatic translation. However, automatic translation still has low accuracy and cannot withstand use in public applications that require high accuracy, such as attracting tourists by local governments.

[0007] Therefore, another object of the present invention is to provide an SNS posting system that can generate a second post content to be posted simultaneously with a first post content in a single operation using highly accurate automatic translation. [Means for solving the problem]

[0008] An SNS posting system according to a first aspect of the present invention is an SNS posting system that generates first post content for uploading to a first SNS account based on user input results, generates second post content for uploading to a second SNS account by modifying the input results according to user settings or operations, and uploads the first post content to the first SNS account and the second post content to the second SNS account according to user operations.

[0009] In the SNS posting system according to the first aspect of the present invention, the first post content may be identical to the input result. Alternatively, the second post content may be generated by automatically translating the input result into a language set by the user, or the second post content may be generated by creating a temporary post content identical to the input result and presenting it to the user, and then the user modifying the presented temporary post content.

[0010] A second aspect of the present invention is an SNS posting system connected to a translation engine used in a media system that has the function of generating and displaying a web page which is an automatic translation of an article using a translation engine configured to accept retraining by a user, wherein the web page is configured to accept reports of translation errors by users viewing the web page and to notify the administrator of the media system of the received reports, and the SNS posting system generates first post content for upload to a first SNS account based on the user's input result, generates second post content for upload to a second SNS account by automatically translating the input result using the translation engine, and uploads the first post content to the first SNS account and the second post content to the second SNS account in response to user operation.

[0011] In the SNS posting system according to the second aspect of the present invention, the first post content may be identical to the input result. The SNS posting system may also generate a temporary post content identical to the input result and present it to the user, and the user may modify the presented temporary post content to generate a third post content for uploading to a third SNS account, and in response to user operations, upload the first post content to the first SNS account, the second post content to the second SNS account, and the third post content to the third SNS account. [Effects of the Invention]

[0012] According to the first aspect of the present invention, a user can generate a second post with content different from the first post without having to input the second post content separately from the first post content. Therefore, according to the first aspect of the present invention, it becomes possible to generate a second post content that is posted simultaneously with the first post content in a single operation with simple operations.

[0013] According to a second aspect of the present invention, a second post, which is a translation of the first post, is generated using a translation engine used in a media system that can improve the accuracy of its translation model based on the identification of translation errors in articles widely published to the public. Therefore, it becomes possible to generate a second post, which is posted simultaneously with the first post, using highly accurate automatic translation in a single operation. [Brief explanation of the drawing]

[0014] [Figure 1] This figure shows the configuration of a system according to an embodiment of the present invention. [Figure 2] This diagram shows the structure of the article table contained in database 13. [Figure 3] This figure shows an example of a web page generated by web server 11 based on a Japanese translation of an article (i.e., the original text itself) when the original text is written in Japanese. [Figure 4] This figure shows an example of a web page generated by the web server 11 based on an English translation of an article (automatically translated using translation engine 3a) when the original text is written in Japanese. [Figure 5] (a) is a diagram showing the structure of the SNS table included in database 13, and (b) is a diagram showing the structure of the posts table included in database 13. [Figure 6] This figure shows the basic hardware configuration of each computer shown in Figure 1. [Figure 7] This diagram shows the transmission and reception of data between the computers and databases that constitute SNS posting system 1 and media system 2. [Figure 8] This is a diagram showing a translation error reporting form. [Figure 9] This is a diagram showing the SNS integration management screen. [Figure 10] This is a diagram showing the article input screen. [Figure 11]It is a diagram showing a child screen displayed when the user presses the reflection button 41 on the article input screen. [Figure 12] It is a diagram showing the posting content set by the management server 10 in the form of tab 39b based on the article input in tab 39a shown in FIG. 10. [Figure 13] It is a diagram showing the posting content set by the management server 10 in the form of tab 39c based on the article input in tab 39a shown in FIG. 10. [Figure 14] It is a diagram showing the posting content set by the management server 10 in the form of tab 39d based on the article input in tab 39a shown in FIG. 10. [Figure 15] It is a diagram showing the state where the posting content shown in FIG. 12 is displayed on the SNS. [Figure 16] It is a diagram showing the state where the posting content shown in FIG. 13 is displayed on the SNS. [Figure 17] It is a diagram showing the state where the posting content shown in FIG. 14 is displayed on the SNS. [Figure 18] It is a diagram showing the sequence of processes executed by each computer shown in FIG. 1. [Figure 19] It is a diagram showing the sequence of processes executed by each computer shown in FIG. 1. [Figure 20] It is a diagram showing the sequence of processes executed by each computer shown in FIG. 1.

Embodiments for Carrying Out the Invention

[0015] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0016] Figure 1 shows the configuration of a system according to an embodiment of the present invention. As shown in the figure, this system has a configuration in which an SNS posting system 1, a media system 2, a translation engine 3, multiple SNS API servers 4 including SNS API servers 4a and 4b, a translation application API server 5, and multiple user terminals 6 including user terminals 6a and 6b are interconnected via a network 7. Of these, the SNS posting system 1 is composed of a management server 10, an API server 12, and a database 13. The media system 2 is composed of a web server 11, an API server 12, and a database 13. The API server 12 and the database 13 are common to both the SNS posting system 1 and the media system 2.

[0017] The SNS posting system 1 is a system that provides a user terminal 6 with the functionality to generate content to be posted to multiple SNS accounts based on user input and to actually post the content. On the other hand, the media system 2 is a system that generates a media site, such as a social video curation site, and provides it to the user terminal 6 via the network 7. Each media site contains numerous web pages, each containing articles, and the SNS posting system 1 is configured to generate content to be posted to multiple SNS accounts based on articles entered by the user for publication on the media site, through user operation. The following describes each configuration in detail.

[0018] First, the translation engine API server 3 is a server computer that provides the function of translating and outputting input text. The translation engine API server 3 has a built-in translation engine 3a and is configured to translate input text using this translation engine 3a.

[0019] The translation engine 3a is configured to understand and generate language using artificial intelligence (AI). Specifically, it is preferable to configure the translation engine 3a with a deep learning or transformer-based language model trained using a wide range of language data. Specific examples of translation engine 3a include Microsoft® Azure® AI Translator and OpenAI® ChatGPT®.

[0020] The translation engine API server 3 includes an interface for training the translation engine 3a, and the translation engine 3a is configured to accept retraining by users by accessing it from the outside via this interface. In this embodiment, this function is used by the administrator of the media system 2 to retrain the translation engine and thereby improve or optimize the accuracy of translations.

[0021] Next, focusing on media system 2, database 13 is a database containing an article table for storing articles from media sites. Articles may include text, images, videos, URL links, and embedded content (such as maps, videos, and posts from other social media platforms). The article table is updated by API server 12.

[0022] Figure 2 shows the structure of the article table included in database 13. As shown in the figure, each record in the article table includes the original article text entered by the user, the type of language in which it is written (Japanese, English, etc.), and the translated article, which is created by translating the original article text into Japanese, English, Traditional Chinese, Simplified Chinese, and Korean, respectively. When the language of the translated article is the same as the language of the original text, it is preferable to use the original text itself for the translated article. In this embodiment, we take an example using a language set consisting of five languages: Japanese, English, Traditional Chinese, Simplified Chinese, and Korean, but the present invention is also suitably applicable when using other language sets. Translation from the original article text to the translated article is performed using the translation engine 3a built into the translation engine API server 3. Details of this will be described later.

[0023] Returning to Figure 1, the API server 12 is a server computer that acts as an interface between the database 13 and other computers. The API server 12 also receives article input from the management server 10, generates translated articles using automatic translation by the translation engine 3a, and stores them in the article table.

[0024] The web server 11 is a server computer that, in accordance with HTTP (Hypertext Transfer Protocol), provides web pages containing HTML (HyperText Markup Language) and objects such as images to browsers running on user terminals 6 and other devices. The web server 11 plays the role of providing a media site to users by reading one or more articles from the article table in the database 13 via the API server 12 and embedding and displaying them within the web page.

[0025] As explained with reference to Figure 2, each record in the article table contains translated articles in Japanese, English, Traditional Chinese, Simplified Chinese, and Korean. The web server 11 is configured to generate a web page with a different URL assigned to each of these translated articles and publish them within the media site. The different URLs may include a directory name indicating the language of the article, such as "··· / jp / ···" or "··· / en / ···".

[0026] Figures 3 and 4 show examples of web pages generated by the web server 11 based on a Japanese translated article (i.e., the original text itself) and an English translated article (an automatic translation of the original text using a translation engine), respectively, when the original text is written in Japanese. Referring to Figure 4, this web page is configured to display, depending on user interaction, the original text 30 (written in Japanese) and a translation error reporting link 31 at the end of the English article. By displaying the original text 30 in this way, users viewing this web page can understand that the displayed English article is an automatic translation. The translation error reporting link 31 is always displayed by the web server 11 along with the display of the original text 30 and is a hyperlink to a translation error reporting form for receiving translation error reports. Details of this will be described later.

[0027] Here, the web pages in each language generated by the web server 11 are part of a media site provided by the web server 11 to the browser of the user terminal 6. When displaying a web page containing an article, this media site is configured to allow the user to select one of the following languages ​​according to the browser's language settings: Japanese, English, Traditional Chinese, Simplified Chinese, or Korean, and then display a web page containing the translated article in the selected language. In this way, the user can view the web page in the language they have set in their browser.

[0028] Furthermore, if a translation error reporting link 31 is posted on the web page as described above, the administrator of media system 2 will be responsible for reviewing the reported translation errors and, if necessary, retraining the translation engine 3a. Therefore, since the administrator himself must understand the language, the types of languages ​​in which translated articles should be created (i.e., the types of languages ​​that make up the above language set) will be determined by the personnel available to the administrator. In the following, the languages ​​that make up the above language set and are used to create web pages by the web server 11 will be referred to as "original translation languages," and other languages ​​will be referred to as "application translation languages."

[0029] Returning to Figure 1, we now focus on the SNS posting system 1. The management server 10 is a server computer that inputs articles to be published on the media site and provides a user interface for generating posts to one or more SNS accounts based on the input articles. This user interface may be a website or an application used on a smartphone or the like.

[0030] Database 13 includes an SNS table that stores information about the SNS accounts to which posts generated by the management server 10 are posted, and a post table that stores the content of posts to each SNS account. Updates to these tables are also performed by the API server 12.

[0031] Figure 5(a) shows the structure of the SNS table included in database 13. As shown in the figure, the SNS table is a table that stores account information including username and password for each SNS account, the type of SNS (Facebook®, Instagram®, X (formerly Twitter®), etc.), and the language type indicating the language used for posting.

[0032] Figure 5(b) shows the structure of the post table included in database 13. As shown in the figure, each record in the post table consists of information about the SNS account to which the post is posted, the post content, the posting date and time, and engagement information. The post content is generated by the management server 10 from the article to be published on the media site. The posting date and time is information indicating the date and time when the corresponding post content was actually posted to the SNS account. The engagement information is information indicating the number of times the corresponding post has been "liked" (including clicks and taps; the same applies hereinafter) or "reposted," and the management server 10 is configured to obtain the engagement information for each post from the SNS_API server 4 and store it in the post table.

[0033] Returning to Figure 1, the SNS_API Server 4 is a server computer that provides an interface for SNS services, enabling interaction and information sharing with others over the internet. Specific examples of SNS services include Facebook®, Instagram®, and X.

[0034] The translation app API server 5 is a server computer that provides an interface to a translation app, which is a SaaS (Software as a Service) for translating text written in one language into another language. A specific example of a translation app is Google Translate®. The browser running on the user terminal 6 uses the translation app via the translation app API server 5 to translate and display the text contained in any web page according to the browser's language settings and user settings.

[0035] Referring again to Figure 3, the web page shown in the figure contains an article in Japanese. If the user wants to view it in the app's translation language (for example, French), the user selects that language in a combo box (not shown). The web server 11 then first redirects the user to a web page containing the corresponding English article, and then accesses the translation app API server 5 to automatically translate the article on the web page into French using the translation app. In this way, the user can read the article in the app's translation language. The reason for the initial redirection to a web page containing the English article is that the translation accuracy of the translation app is expected to be higher for English-to-French translation than for Japanese-to-French translation. Depending on the target language, a language other than English may be used as an intermediary. Also, the article displayed in the app's translation language in this way does not include the translation error reporting link 31 shown in Figure 4.

[0036] Returning to Figure 1, the user terminal 6 is a personal computer, and can be comprised of, for example, a desktop or notebook computer, a tablet, or a smartphone. In this embodiment, the user terminal 6 is used to access the internet via a browser, to view web pages provided by the web server 11, and to perform various inputs to the management server 10 and the translation engine 3a.

[0037] Network 7 is, for example, the Internet, and it plays the role of connecting each of the computers shown in Figure 1 and enabling the transmission and reception of information between them.

[0038] Figure 6 shows the basic hardware configuration of each computer shown in Figure 1. As shown in the figure, each device consists of a computer 100 having a configuration in which a processor 101, a storage device 102, a communication device 103, an input device 104, and an output device 105 are interconnected via a bus 106.

[0039] The processor 101 is a central processing unit that reads and executes programs stored in the storage device 102. In this embodiment, each process described as being performed by each computer is realized by the processor 101 of each computer reading and executing programs stored in its respective storage device 102. The processor 101 is configured to communicate with various parts within the computer via the bus 106, and performs tasks such as controlling various parts and processing data stored in the storage device 102, according to the description of the program to be executed.

[0040] The storage device 102 is a device that temporarily or permanently stores various programs and data. The storage device 102 is usually composed of a combination of multiple storage devices, such as a main memory device consisting of DRAM (Dynamic Random Access Memory) or SRAM (Static Random Access Memory), and an auxiliary storage device consisting of a hard disk or SSD (Solid State Drive).

[0041] The communication device 103 is a device that performs communication with an external computer (including the network 7 shown in Figure 1) in response to the control of the processor 101. The method of communication performed by the communication device 103 is not particularly limited, but examples include wired or wireless WAN (Wide Area Network) or LAN (Local Area Network).

[0042] The input device 104 is a device that receives input from the user and includes various input means such as a mouse, keyboard, and touch panel. The content of the user input received by the input device 104 is transmitted to the processor 101 via the bus 106. The output device 105 is a device that provides output to the user in accordance with the control of the processor 101 and includes various output means such as a display and speakers.

[0043] The functions of SNS posting system 1 and media system 2 will be explained in detail below, with reference to Figures 7 to 20.

[0044] Figure 7 shows the transmission and reception of data between the computers and databases that constitute the SNS posting system 1 and the media system 2. Focusing on the media system 2 first, the API server 12 receives input of a new article from the user terminal 6 via the management server 10 and processes data to be stored in the article table in the database 13.

[0045] Specifically, the API server 12 determines the language type of the article input from the user terminal 6 and uses the translation engine 3a to automatically translate it into each original translation language. The API server 12 then stores the original article, the determined language type, and each translated article obtained through automatic translation into the article table in the database 13, as explained with reference to Figure 2. As a result of this storage, the web server 11 publishes a web page containing the article written in each original translation language as part of the media site on the internet.

[0046] The web server 11 also processes reports of translation errors in translated articles from users via the translation error reporting link 31, as explained with reference to Figure 4. Specifically, when a user clicks the translation error reporting link 31, a predetermined translation error reporting form is displayed, and the system is configured to accept translation error reports through this form.

[0047] Figure 8 shows a translation error reporting form. The translation error reporting form includes an input field 32 for the translation to be corrected, an input field 33 for the correct translation, an input field 34 for an optional comment, and a submit button 35. When a user enters information into input fields 32-34 and presses the submit button 35, the content entered into each input field 32-34 is supplied to the web server 11 from the browser running on the user terminal 6. The web server 11 then notifies the administrator of the media system 2 of the supplied content via email or other means. Upon receiving this notification, the administrator reviews the validity of the user's input and, if deemed necessary, retrains the translation engine 3a. This reduces the possibility of similar translation errors occurring.

[0048] The reason why the process between reporting a translation error and retraining the translation engine 3a involves a decision by the administrator is that if we were to accept reports of errors from the general public who read the article, some of the reports might not be valid. If the report is deemed valid, such as when limiting who can report errors, the web server 11 may directly access the translation engine API server 3 to retrain the translation engine 3a.

[0049] Returning to Figure 7, the management server 10 then generates content to be posted to each SNS based on the article (article to be posted to the media system 2) entered from the user terminal 6, and stores it in the posting table shown in Figure 5(b) via the API server 12. The management server 10 also receives information about the SNS account to which the post will be made (including the type of language to be posted) from the user terminal 6, and stores it in the SNS table shown in Figure 5(a) via the API server 12.

[0050] Figure 9 shows the SNS integration management screen that the management server 10 displays in the browser of the user terminal 6 to receive SNS account information. The management server 10 is configured to configure the SNS table in response to user operations on this SNS integration management screen.

[0051] As shown in Figure 9, the SNS integration management screen is configured to have a connection button 36 and one or more rows corresponding to SNS accounts for each type of SNS. In the example in Figure 9, the SNS integration management screen has rows corresponding to account "Cjvfb02" and account "Cjvfb01" for Facebook (registered trademark), and a row corresponding to account "CJV_X_02" for X. Each row has a language setting combo box 37 and a trash can button 38.

[0052] On the SNS integration management screen, when a user clicks, for example, the Facebook® connection button 36, the Facebook® login screen opens. When the user logs in to Facebook® on this login screen, the management server 10 retrieves information about the SNS account used for that login. If this information is not already set in the SNS table, it adds a record of that information to the SNS table and displays a new row on the SNS integration management screen. The same process applies to other SNS accounts.

[0053] The language setting combo box 37 is configured to allow the user to select one of several original languages. When a user selects an original language in this language setting combo box 37, the management server 10 stores the language type indicating that original language in the corresponding record in the SNS table. Furthermore, when the user presses the trash can button 38, the management server 10 deletes the corresponding record from the SNS table and removes the corresponding row from the SNS integration management screen.

[0054] Returning to Figure 7, the management server 10 is configured to generate an article input screen for receiving article input and display it in the browser of the user terminal 6. On this article input screen, the user enters the article and publishes it to the media site, and also processes the posting of the content generated based on the article to each SNS configured in the SNS table.

[0055] Figure 10 shows the article input screen. As shown in the figure, this article input screen consists of multiple tabs 39, including tabs 39a to 39d, and a post button 40. In addition, a reflect button 41 is located within tab 39a. The reflect button 41 is not located in any of the other tabs 39.

[0056] Tab 39a is a page for accepting article input, and the articles entered here are supplied to the API server 12 by the management server 10. The processing performed by the API server 12 upon receiving this supply is as described above. The other tabs 39b to 39d are pages that present the user with the content to be posted to each of the multiple SNS accounts set in the SNS table, and allow them to change it as needed. Tabs 39b to 39d correspond to the accounts "Cjvfb02", "Cjvfb01", and "CJV_X_02" shown in Figure 9, respectively.

[0057] When the user presses the reflect button 41 while tab 39a is displayed, the management server 10 generates post content for each SNS based on the article entered in tab 39a and reflects the generated post content in the corresponding form in tab 39. If the article contains images, videos, URL links, embedded content, etc., the management server 10 also reflects these in the form in each tab 39.

[0058] Figure 11 shows a sub-screen that appears when the user presses the Apply button 41. As shown in the figure, this sub-screen has a language setting combo box 42 for each SNS account set in the SNS table, as well as an Apply button 43. The language setting combo box 42 allows the user to re-select the language of posts to the corresponding SNS account, and, similar to the language setting combo box 37 shown in Figure 9, it is configured to allow selection of one of several original languages. The language setting combo box 42 is displayed with a selection made according to the language type in the SNS table.

[0059] When a user selects a language in the language setting combo box 42 and presses the Apply button 43, the management server 10 first updates the language type of each record in the SNS table using the selected language information. Next, the management server 10 generates post content for each SNS account based on the language type of the SNS table that has been updated in this way. For example, if the article entered in tab 39a is in Japanese and the language type of a certain SNS account is Japanese, the post content generated for that SNS account will be the article entered in tab 39a itself. Also, if the article entered in tab 39a is in Japanese and the language type of a certain SNS account is English, the post content generated for that SNS account will be the article entered in tab 39a automatically translated into English by the translation engine 3a. After generating the post content for each SNS account, the management server 10 sets the generated post content in the form in the corresponding tab 39.

[0060] Figures 12 to 14 show the post content that the management server 10 sets in the respective forms of tabs 39b to 39d, based on the article entered in tab 39a shown in Figure 10. In Figure 10, the article entered in tab 39a is written in Japanese, and as shown in Figure 11, the language types set in the SNS table for SNS accounts "Cjvfb02", "Cjvfb01", and "CJV_X_02" are Japanese, English, and Traditional Chinese, respectively. Therefore, as shown in Figures 12 to 14, the forms of tabs 39b to 39d will be set with post content in Japanese, English, and Traditional Chinese, respectively.

[0061] Within tabs 39b to 39d, the content of posts set in the forms, including non-text elements, is configured to be editable by the user. This allows users to customize the content of their posts as needed, for example, to accommodate SNS character limits or cultural differences in different countries. Once all edits are complete and the user presses the post button 40, the management server 10 supplies the article set in tab 39a to the API server 12, and stores the content of posts set in each of the other tabs 39 in the post table via the API server 12. The processing performed by the API server 12 upon receiving the article is as described above. The management server 10 promptly uploads the content of posts stored in the post table to the corresponding SNS accounts. As a result, the content of posts generated based on the article is simultaneously published to each SNS account.

[0062] Here, the article input screen may be configured to allow the user to specify the actual posting date and time to the SNS in each tab 39. In this case, the specified date and time will be set in the posting date and time shown in Figure 5(b), and the management server 10 will upload the corresponding post content when the posting date and time stored in the posting table arrives. This makes it possible for users to generate post content in advance that they plan to post in the future.

[0063] Figures 15 to 17 show how the posted content shown in Figures 12 to 14 appears on social media. Figures 15 and 16 are screenshots from Facebook (registered trademark), and Figure 17 is a screenshot from X. As shown in these figures, the social media posting system 1 allows users to post the same content as articles published in the media system to each social media platform in various languages. Although not shown in Figures 15 to 17, it is also possible to post slightly modified versions of articles published in the media system to each social media platform.

[0064] Figures 18 to 20 show the sequence of processes performed by each computer shown in Figure 1. The processes according to this embodiment will be described in more detail below with reference to these figures.

[0065] Figure 18 shows the sequence of operations for simultaneously posting an article from a media site to multiple social networking services (SNS). As shown in the figure, first, a common input screen (see Figure 10) is sent from the management server 10 to the user terminal 6 (step S1). When the user enters an article in the article input tab (tab 39a shown in Figure 10) of this common input screen and presses the submit button 41 (see Figure 10), the management server 10 first determines the language type of the entered article (step S4). Then, for each SNS account stored in the SNS table shown in Figure 5(a) (step S5), the management server 10 compares its language type with the language type determined in step S4 to determine whether translation is necessary (step S6). If it determines that translation is necessary, it accesses the translation engine API server 3 and performs automatic translation of the article using the translation engine 3a (steps S7, S8). In step S6, the management server 10 determines that for SNS accounts that do not require translation, it sets the original article as the post content in the form, and for SNS accounts that it determines that require translation in step S6, it sets the translated article returned by the translation engine 3a as the post content in the form (step S9).

[0066] If a user modifies the content of a post set in the form (step S10), the management server 10 reflects the modification in the post (step S11). Then, when the user presses the post button 40 (see Figure 10) (step S12), the management server 10 stores the content of the post to each SNS in the post table in the database 13 (see Figure 5(b)) via the API server 12 (step S13). This enables simultaneous posting of articles from a media site to multiple SNSs.

[0067] Figure 19 shows the sequence of processes involved in posting an article to a media site. Steps S1, S2, and S12 shown in the figure are the same processes as those in the sequence shown in Figure 18. When the post button 40 (see Figure 10) is pressed in step S12, the management server 10 supplies the original article text to the API server 12 (step S20). Upon receiving this supply, the API server 12 first determines the type of language in which the input article text is written (step S21). Then, for each of the original languages ​​other than the original text's language, it performs automatic translation of the article using the translation engine 3a (steps S22-S24). Subsequently, the API server 12 stores the original article text, its language type, and each translated article in the article table in the database 13 (see Figure 2). This enables the posting of the article to the media site.

[0068] Figure 20 shows the sequence of operations for receiving reports of translation errors in translated articles published on a media site from users. In this process, first, a web page containing an automatic translation of the article in its original language (see Figure 4) is sent from the web server 11 to the user terminal 6 (step S30). This web page includes a translation error report link 31, as explained with reference to Figure 4. When the user clicks this translation error report link 31 (step S31), the translation error report form shown in Figure 8 is displayed (step S32). When the user fills in the details of the translation error in this translation error report form (step S33) and clicks the submit button 35 (see Figure 8) (step S34), the web server 11 notifies the administrator of the media system 2 of the entered translation error details. This makes it possible to retrain the translation engine 3a based on the reported translation errors.

[0069] As described above, with the SNS posting system 1 according to this embodiment, the user can generate a second post with content different from the first post (for example, the post shown in Figure 13 or Figure 14) without having to input a second post with content different from the first post (for example, the post shown in Figure 12). Therefore, with the SNS posting system 1 according to this embodiment, it becomes possible to generate a second post that is posted simultaneously with the first post with a simple operation in a single operation.

[0070] Furthermore, according to the SNS posting system 1 of this embodiment, the second post content, which is a translation of the first post content, is generated using a translation engine 3a used in a media system 2 that can improve the accuracy of the translation model based on the identification of translation errors in articles that are widely published to the public. Therefore, it becomes possible to generate the second post content, which is posted simultaneously with the first post content, using highly accurate automatic translation in a single operation.

[0071] Although preferred embodiments of the present invention have been described above, the present invention is not limited in any way to these embodiments, and it goes without saying that the present invention can be implemented in various forms without departing from its essence.

[0072] For example, in the above embodiment, an example was described in which content to be posted to each SNS account is generated based on an article published in the media system. However, in addition to the article published in the media system, content to be posted to each SNS account may also be generated from the user's input results. In this generation, the text obtained by inputting the user's input results as a prompt into the generation AI may be used as the content to be posted to each SNS account. In this case as well, according to the present invention, it becomes possible to generate a second post that is posted simultaneously with the first post in a single operation with simple operations.

[0073] Furthermore, while the above embodiment described an example of generating content for multiple SNS accounts from a single article, the present invention is also applicable when generating content for only one SNS account from a single article (or user input). In this case, according to the present invention, it becomes possible to create the content for a post from a single article (or user input) with simple operations. In addition, it becomes possible to generate the content for a post generated from a single article (or user input) using highly accurate automatic translation.

[0074] Furthermore, in the above embodiment, one post is generated in one language, but it may also be generated with multiple languages ​​listed. In this case, the SNS integration management screen shown in Figure 9 can be configured to allow selection of multiple languages ​​for a single SNS account, and the SNS table shown in Figure 5(a) can be configured to store multiple language types for each SNS account. When the management server 10 generates the post, it can automatically translate the article into each of these multiple language types and then set the result of this translation in the corresponding form in tab 39.

[0075] Furthermore, in the above embodiment, we described an example where pressing the post button 40 shown in Figure 10 immediately uploads the post content generated from the article to the SNS account (or at a specified date and time). However, it is also possible to introduce a process to obtain permission from a supervisor before uploading. This would reduce the possibility of the post content causing a social media firestorm. [Explanation of symbols]

[0076] 1. SNS posting system 2 Media Systems 3 Translation Engine API Server 3a Translation engine 4,4a,4b SNS_API Server 5. Translation app API server 6,6a,6b User terminals 7 Network 10 Management Server 11 Web Server 12 API Servers 13 Databases 30 Original text 31 Report translation error link 32-34 Input fields 35 Send button 36. Connect button 37. Language setting combo box 38 Trash button Tabs 39, 39a~39d 40 Post button 41 Apply button 42. Language setting combo box 43 Apply button 100 Computers 101 Processors 102 Storage device 103 Communication equipment 104 Input device 105 Output device 106 Bus

Claims

1. Based on the user's input, generate a first post content to upload to the first SNS account. By modifying the input results according to the user's settings or operations, a second post content for uploading to a second SNS account is generated. In response to user operation, the first post content is uploaded to the first SNS account, and the second post content is uploaded to the second SNS account. SNS posting system.

2. The content of the first post mentioned above is identical to the input result mentioned above. The SNS posting system according to claim 1.

3. The second post content is generated by automatically translating the input results into the language set by the user. The SNS posting system according to claim 1.

4. The second post content is generated by creating a provisional post content identical to the input result and presenting it to the user, and then the user modifies the presented provisional post content. The SNS posting system according to claim 1.

Images