Method for automatically translating blog text based on generative machine learning model

Abstract

The present invention relates to a method for providing a service by translating a post of a blog in a plurality of languages. An embodiment of the present invention relates to a method for translating summary text of a blog on the basis of a generative machine learning model. The method comprises the steps in which: a first computing device generates summarized text by inputting post data to the generative machine learning model; the first computing device transmits the summarized text to a plurality of second computing devices; and each of the plurality of second computing devices generates summary text translated in different languages by inputting the summarized text to the generative machine learning model.

Description

A method for automatically translating blog text based on a generative machine learning model.

[0001] The present invention relates to a method for providing a service by translating a blog post into multiple languages.

[0002] More specifically, it relates to a technology that translates the content of a post, summarizes the post, and keywords extracted from the post into multiple languages, selects a category for the post, and evaluates the quality of the post to determine the level of exposure.

[0003] A blog is a service that collects and posts posts written for the purpose of sharing information or exchanging opinions. The word "blog" is a combination of "web" and "log." It's a form of writing where the writer individually records and posts his or her thoughts, feelings, and opinions about people, objects, events, natural phenomena, and more. Blogs are publicly accessible on the web, allowing others to read and share these posts.

[0004] Blogs can be categorized into service-based blogs offered by portal sites or web service providers and accessible through membership registration and application; installable blogs, which can be installed on users' devices or resources provided by cloud services; and microblogs with restrictions on the number of characters per post or the capacity of multimedia content. Companies operating blog services offer various incentives, such as advertising revenue sharing, to encourage users to post on their services.

[0005] However, the blogging services and programs currently available to users require users to post in their native language or a language they speak, and only those who can read that language can understand the content. Therefore, posts written in a specific language cannot be accessed and enjoyed in cultures that speak other languages, hindering the dissemination of high-quality content.

[0006] To address this, translating and reprocessing posts into different languages ​​can be incredibly time-consuming and costly.

[0007] The purpose of the present invention is to solve the above problems and provide a method for translating blog posts into multiple languages ​​and providing services.

[0008] It also provides a method to summarize posts, extract keywords, translate them into multiple languages, select categories for posts, and evaluate the quality of posts to determine their exposure.

[0009] In order to achieve the above object, one embodiment of the present invention provides a method for translating a summary text of a blog based on a generative machine learning model, the method comprising: a step in which a first computing device inputs post data into a generative machine learning model to generate a summarized text; a step in which the first computing device transmits the summarized text to a plurality of second computing devices; and a step in which the plurality of second computing devices each input the summarized text into the generative machine learning model to generate a summarized text translated into different languages.

[0010] In one embodiment, the method further includes the step of transmitting the translated summary text to the terminal device of the second user when the plurality of second computing devices receive a message requesting retrieval of the summarized text from the terminal device of the second user.

[0011] In one embodiment, the post data received from the terminal device of the first user includes text of the title, body, and subtitle.

[0012] In one embodiment, the method further includes a step of a first computing device inputting post data into a generative machine learning model to select a category; and a step of the first computing device setting the selected category as a category of the post data.

[0013] In one embodiment, the method further includes: a step of inputting post data into a generative machine learning model by a first computing device to calculate a score; and a step of exposing the post data if the score of the post data is equal to or greater than a reference score, and not exposing the post data if the score is less than the reference score.

[0014] The present invention allows users of a blog service to write a post and register it in the system, and since the post is automatically translated into each language and published, even if other users do not know the language of the post, they can easily understand the content of the post through the translated post.

[0015] Additionally, by displaying the summarized text, keywords, and categories of the post to users, you can easily understand the type and content of the post, and by determining whether to expose the post based on its score, you can ensure that high-quality content is used.

[0016] FIG. 1 is a drawing briefly showing a multilingual automatic translation blog service system and other components according to one embodiment of the present invention.

[0017] FIG. 2 is a diagram schematically illustrating a configuration included in a multilingual automatic translation blog service system according to one embodiment of the present invention.

[0018] FIG. 3a and FIG. 3b are schematic drawings showing a post registration interface and a post search interface, respectively, in one embodiment of the present invention.

[0019] FIG. 4 is a diagram illustrating a process of translating a registered post in one embodiment of the present invention.

[0020] FIG. 5 is a diagram illustrating a process of summarizing registered posts in one embodiment of the present invention.

[0021] FIG. 6 is a diagram illustrating a process of extracting keywords from a registered post in one embodiment of the present invention.

[0022] FIG. 7 is a diagram illustrating a process for selecting a category of a registered post in one embodiment of the present invention.

[0023] FIG. 8 is a diagram illustrating a process for calculating a score of a registered post in one embodiment of the present invention.

[0024] FIG. 9 is a diagram briefly illustrating a multilingual automatic translation blog service method according to one embodiment of the present invention.

[0025] FIG. 10 is a diagram schematically showing the configuration of a computing device in one embodiment of the present invention.

[0026] The following merely illustrates the principles of the present invention. Therefore, those skilled in the art will be able to implement the principles of the present invention and invent various devices within the scope and spirit of the present invention, even if not explicitly described or illustrated herein. Furthermore, all conditional terms and embodiments listed herein are expressly intended, in principle, to facilitate understanding of the concepts of the present invention, and should be understood as being in no way limited to the specifically listed embodiments and conditions.

[0027] For example, throughout the specification, when a part is said to be "connected" to another part, this includes not only cases where it is "directly connected" but also cases where it is "indirectly connected" with other elements intervening. Furthermore, when a part is said to "include" a component, this does not exclude other components, but rather implies that it may include other components, unless otherwise specifically stated.

[0028] Furthermore, it should be understood that all detailed descriptions of the principles, aspects, and embodiments of the present invention, as well as specific embodiments, are intended to encompass structural and functional equivalents thereof. Furthermore, it should be understood that such equivalents encompass not only currently known equivalents but also equivalents developed in the future, i.e., all devices invented to perform the same function, regardless of structure.

[0029] The above-described purposes, features, and advantages will become more apparent through the following detailed description, taken in conjunction with the accompanying drawings. Accordingly, those skilled in the art will be able to readily implement the technical concepts of the present invention. Furthermore, in describing the present invention, if a detailed description of known technology related to the present invention is deemed to unnecessarily obscure the gist of the invention, such a detailed description will be omitted.

[0030] FIG. 1 is a drawing briefly showing a multilingual automatic translation blog service system and other components according to one embodiment of the present invention.

[0031] A multilingual automatic translation blog service system (100, hereinafter referred to as the "system") according to one embodiment of the present invention may be a single computing device or a collection of multiple computing devices interconnected via a computer network. Each computing device may include a processor, a memory device, and an input / output device, the configuration of which is described in detail in FIG. 10.

[0032] The first user may be a person who registers a post in the system (100). The post may be multimedia content such as text and static images, videos, and sounds containing text.

[0033] A second user may be someone who views and uses posts registered by the first user. The second user may read the posts registered by the first user or share them with other users.

[0034] The first user and the second user can register or view posts using terminal devices (A, B), respectively.

[0035] The user terminal devices (A, B) may be computing devices having a processor, memory, and input / output devices. The input / output devices may include a communication module capable of connecting to a computer network. The communication module may be, for example, one or more of an Ethernet card, which is a network interface controller capable of connecting to a wired local area network (LAN), a Wi-Fi module capable of connecting to a wireless local area network (WLAN), a Bluetooth module, and a cellular modem capable of transmitting and receiving data via a mobile network. The user terminal devices (A, B) may be, for example, any one of a desktop computer, a laptop computer, a tablet computer, a smartphone, and a smartwatch, but are not limited to a specific type of device.

[0036] FIG. 2 is a diagram schematically illustrating a configuration included in a multilingual automatic translation blog service system according to one embodiment of the present invention.

[0037] A multilingual automatic translation blog service system (100) according to one embodiment of the present invention may include a post management module (110), a translation module (120), a post summary module (130), a keyword extraction module (140), a category selection module (150), and a quality measurement module (160).

[0038] Each of the modules (110 to 160) may be a computing device, or all or part of the modules (110 to 160) may be application software running on a computing device. For example, a post management module (110), a post summary module (130), a keyword extraction module (140), a category selection module (150), and a quality measurement module (160) may be run as application software on a first computing device, and a translation module (120) may be run as application software on a plurality of second computing devices to translate posts into different languages, respectively.

[0039] The post management module (110) can store, edit, and delete posts created by the first user. The post management module (110) can provide general blog service functions, such as allowing users to register and edit posts on a blog or to view posts registered by other users.

[0040] The post management module (110) receives a message requesting registration of a post and the post data from the terminal device (A) of the first user, and then stores the message as a file in a storage device or builds a database and adds it as a record to a table.

[0041] The post management module (110) receives a message requesting modification of a post and data including the modification details of the post from the terminal device (A) of the first user, and then changes the file storing the post in response to the modification details or changes a record in a table of a database.

[0042] The post management module (110) receives a message requesting deletion of a post from the terminal device (A) of the first user, and then deletes the file storing the post or removes a record from a table in the database.

[0043] The post management module (110) can receive a message requesting a post search from a terminal device (B) of a second user, and then retrieve post data from a table in a file or database and transmit it to the terminal device (B) of the second user.

[0044] The translation module (120) can translate posts, summarized posts, and extracted keywords into multiple languages.

[0045] The translation module (120) can execute translation tasks using a first generative machine learning model. The first generative machine learning model may be trained to translate a post written in one language into multiple other languages. Furthermore, the first generative machine learning model may be trained to translate summarized posts and extracted keywords into multiple other languages.

[0046] The translation module (120) and the first generative machine learning model may be included and executed in a single first computing device. Alternatively, the first generative machine learning model may be included and executed in a single first computing device, and a plurality of translation modules (120) may be included and executed in a plurality of second computing devices, respectively. Alternatively, a plurality of translation modules (120) and a plurality of first generative machine learning models may be included and executed in a plurality of second computing devices, respectively.

[0047] The post written by the first user, the summarized text of the post, and the keywords extracted from the post can be translated by the translation module (120) and then transmitted to the terminal device (B) of the second user in response to the inquiry request of the second user.

[0048] The post summary module (130) can generate a summarized text of a post written by a first user.

[0049] The post summary module (130) can summarize posts using a second generative machine learning model. The second generative machine learning model may be trained to generate summarized text from posts.

[0050] The keyword extraction module (140) can extract multiple keywords corresponding to the content of a post written by the first user.

[0051] Keywords can be words related to the post's topic, expressed in hashtag format. Keywords can also include a summary of the post's text.

[0052] The keyword extraction module (140) can extract multiple keywords from a post using a third generative machine learning model. The third generative machine learning model may be trained to extract multiple keywords from a post.

[0053] The category selection module (150) can select a category corresponding to the content of a post written by the first user.

[0054] The category selection module (150) can select categories from posts using a fourth generative machine learning model. The fourth generative machine learning model may be trained to select categories from posts.

[0055] The quality measurement module (160) can calculate the score of a post written by a first user, and the post management module (110) can determine whether to display the post based on the calculated score.

[0056] The quality measurement module (160) can calculate a score for a post using a fifth generative machine learning model. The fifth generative machine learning model may be trained to calculate a score for a post.

[0057] The first through fifth generative machine learning models may be configured individually, but a single generative machine learning model can execute the functions of the first through fifth generative machine learning models. That is, a single generative machine learning model can translate a post written in one language into multiple other languages, summarize the post, extract multiple keywords from the post, select categories, and calculate a post score on a single computing device or a set of multiple computing devices connected to each other via a computer network.

[0058] FIG. 3a and FIG. 3b are schematic drawings showing a post registration interface and a post search interface, respectively, in one embodiment of the present invention.

[0059] A first user can run application software on a terminal device (A) and request the post management module (110) to create a new post. The post management module (110) transmits a web page having a post registration interface to the terminal device (A), and the terminal device (A) displays the received web page to allow the first user to create a post.

[0060] The post registration interface may display a form for entering a title and body text. Optionally, a form for entering a subtitle may be displayed.

[0061] When a first user inputs a title, body, and optionally a subtitle, and then selects the registration menu, the post entered by the first user can be stored in the system (100) in the same language. The system (100) transmits the post to multiple computing devices, and the multiple computing devices can each translate and store the post in different languages.

[0062] The system (100) can generate summarized text from posts, extract keywords, and select categories. Furthermore, the system (100) can calculate a quality score for the posts.

[0063] A second user can request to view a post registered or translated in the system (100) by executing application software on a terminal device (B). The system (100) transmits a web page having a post viewing interface displaying the title, body, subtitle, summarized text, extracted keywords, and selected categories of the post to the terminal device (B), and the terminal device (B) displays the received web page to allow the second user to view the post.

[0064] A second user can request the system (100) to view a post registered or translated in his / her native language or a language he / she can speak, and the system (100) can transmit the registered post or translated post corresponding to the language selected by the second user to the terminal device (B).

[0065] FIG. 4 is a diagram illustrating a process of translating a registered post in one embodiment of the present invention.

[0066] When the first user inputs a post (title, body, subtitle, etc.) on the terminal device (A) and then selects the registration menu, the terminal device (A) can transmit the post data to the post management module (110) running on the first computing device (C1).

[0067] The post management module (110) can verify the received post data and then save it as a file or add it as a record to a table of the first database. Verification of the post data can include, for example, an integrity check task to check whether the post data transmitted by the terminal device (A) is the same as the post data received by the post management module (110), and a validity check task to check whether the post data includes a code outside the valid range. The table of the first database can include columns such as an identification number that distinguishes the post, an identification number that distinguishes the first user who is the author, the author's name, the title, body, and subtitle text of the post, and the time the post was registered. Each record or row of the table can represent each post.

[0068] Once the verification process is complete, the post management module (110) can publish the registered post so that the second user can view it. That is, when the post management module (110) running on the first computing device (C1) receives a message requesting a post view from the second user's terminal device (B), it can retrieve the post data from a file or a table of the first database and transmit it to the second user's terminal device (B).

[0069] The post management module (110) can then transmit the data of the registered post to each of the plurality of second computing devices (C2). The translation module (120) running on the plurality of second computing devices (C2) can input the received post data into the first generative machine learning model included in the second computing devices (C2) to generate post data translated into different languages. Each of the second computing devices (C2) can translate the posts registered by the first user into different languages ​​(e.g., English, Chinese, Japanese, French, Spanish, etc.).

[0070] The translation module (120) can then store the translated post data as a file in the memory of the second computing device (C2) or add it as a record to a table in a second database. The table in the second database can include columns such as an identification number that distinguishes the translated post, an identification number of the post stored in the first database, the title, body, and subtitle text of the translated post, and the time the post was translated. Each record or row in the table can represent each translated post.

[0071] In contrast, a translation module (120) running on a first computing device (C1) can input data of a registered post into a generative machine learning model included in the first computing device to generate post data translated into different languages, and distribute the post data translated into different languages ​​to a second computing device (C2) corresponding to the translated language.

[0072] Once the distribution task is completed, the post management module (110) can publish the translated post so that the second user can view it. That is, when the post management module (110) running on the second computing device (C2) receives a message requesting a post view from the second user's terminal device (B), it can retrieve the data of the translated post from a file or a table of the second database and transmit it to the second user's terminal device (B).

[0073] FIG. 5 is a diagram illustrating a process of summarizing registered posts in one embodiment of the present invention.

[0074] The first computing device (C1) receives and verifies data of a post entered by a first user, stores the post data, publishes the data so that a second user can view it, and then the post summary module (130) inputs the registered post data into a second generative machine learning model included in the first computing device (C1) to generate summarized text.

[0075] The post summary module (130) can save the summarized text as a file or add it as a column to a record corresponding to a registered post in a table of the first database. Additionally, the table of the first database can include the summarized text as a column.

[0076] The post summary module (130) can then transmit the summarized text to a plurality of second computing devices (C2), respectively. The translation module (120) running on the plurality of second computing devices (C2) can input the summarized text into a first generative machine learning model included in the second computing devices (C2), thereby generating summarized text translated into different languages. Each second computing device (C2) can translate the summarized text into a different language (e.g., English, Chinese, Japanese, French, Spanish, etc.).

[0077] The translation module (120) can then store the translated summary text as a file in the memory of the second computing device (C2), or add it as a column to a record corresponding to the translated post in a table of the second database. Additionally, the table of the second database can include the translated summary text as a column.

[0078] In contrast, a translation module (120) running on a first computing device (C1) can input summarized text into a generative machine learning model included in the first computing device (C1), thereby generating summarized text translated into different languages, and can distribute the summarized text translated into different languages ​​to a second computing device (C2) corresponding to the translated language.

[0079] Once the distribution task is completed, the post management module (110) can publish the translated summary text so that the second user can view it. That is, when the post management module (110) running on the second computing device (C2) receives a message requesting a view of the summarized text from the second user's terminal device (B), it can retrieve the translated summary text from a file or a table in the second database and transmit it to the second user's terminal device (B).

[0080] FIG. 6 is a diagram illustrating a process of extracting keywords from a registered post in one embodiment of the present invention.

[0081] The first computing device (C1) receives and verifies data of a post entered by a first user, stores the post data, publishes the data so that a second user can view it, and then the keyword extraction module (140) inputs the registered post data into a third generative machine learning model included in the first computing device (C1) to extract keywords from the registered post. The extracted keywords may include text summarized from the registered post.

[0082] The keyword extraction module (140) can store the extracted keywords as a file or add them as columns to records corresponding to registered posts in a table of the first database. Additionally, the table of the first database can include the extracted keywords as columns. The keyword extraction module (140) can then transmit the extracted keywords to each of the plurality of second computing devices (C2). The translation module (120) executed in the plurality of second computing devices (C2) can input the extracted keywords into the first generative machine learning model included in the second computing devices (C2) to obtain extracted keywords translated into different languages. Each of the second computing devices (C2) can translate the extracted keywords into different languages ​​(e.g., English, Chinese, Japanese, French, Spanish, etc.).

[0083] The translation module (120) can then store the translated extracted keywords as a file in the memory of the second computing device (C2) or add them as columns to records corresponding to the translated posts in a table of the second database. Additionally, the table of the second database can include the translated extracted keywords as columns.

[0084] In contrast, a translation module (120) running on a first computing device (C1) can input extracted keywords into a generative machine learning model included in the first computing device (C1) to obtain extracted keywords translated into different languages, and distribute the extracted keywords translated into different languages ​​to a second computing device (C2) corresponding to the translated language.

[0085] Once the distribution task is completed, the post management module (110) can publish the translated extracted keywords so that the second user can view them. That is, the post management module (110) running on the second computing device (C2) can receive a message requesting a view of the extracted keywords from the second user's terminal device (B), and then retrieve the translated extracted keywords from a file or a table of the second database and transmit them to the second user's terminal device (B).

[0086] FIG. 7 is a diagram illustrating a process for selecting a category of a registered post in one embodiment of the present invention.

[0087] The first computing device (C1) receives and verifies the data of a post entered by a first user, stores the post data, and publishes it so that a second user can view it. Then, the category selection module (150) inputs the registered post data and category list data into a fourth generative machine learning model included in the first computing device (C1) to select a category of the registered post.

[0088] The category selection module (150) can set the selected category as a category of a registered post by storing the selected category as a file in the memory of the first computing device (C1) or adding it as a column to a record corresponding to the registered post in a table of the first database. Additionally, the table of the first database can include the selected category as a column.

[0089] The post management module (110) can allow a second user to view the categories of registered posts. That is, the post management module (110) running on the first computing device (C1) can receive a message requesting to view the categories of registered posts from the terminal device (B) of the second user, and then retrieve the categories of registered posts from a file or a table of the first database and transmit them to the terminal device (B) of the second user.

[0090] FIG. 8 is a diagram illustrating a process for calculating a score of a registered post in one embodiment of the present invention.

[0091] The first computing device (C1) receives and verifies data of a post entered by a first user, stores the post data, and publishes the data so that a second user can view it. Then, the quality measurement module (160) inputs the registered post data into a fifth generative machine learning model included in the first computing device (C1) to calculate a score of the registered post.

[0092] The quality measurement module (160) can store the calculated score as a file in the memory of the first computing device (C1), or add it as a column to a record corresponding to the registered post in a table of the first database, thereby setting it as a category of the registered post. Additionally, the table of the first database can include the calculated score as a column.

[0093] The post management module (110) can set exposure settings so that a second user can view the registered post if the score of the registered post data is equal to or higher than a reference score, and can set exposure prohibition settings so that the second user cannot view the registered post if the score is lower than the reference score. The second user can only request viewing of posts that have been set to be displayed on the terminal device (B).

[0094] FIG. 9 is a diagram briefly illustrating a multilingual automatic translation blog service method according to one embodiment of the present invention.

[0095] First, the post management module (110) receives and verifies post data, and then registers and publishes the post (step 1010).

[0096] When a first user inputs a post (title, body, subtitle, etc.) into a terminal device (A) and then selects a registration menu, the terminal device (A) can transmit the post data to a post management module (110) running on a first computing device (C1).

[0097] The post management module (110) can execute a verification task to confirm the integrity and validity of the received post data, and then save the received post data as a file or add it as a record to a table of the first database.

[0098] The post management module (110) can publish registered posts so that second users can view them.

[0099] Next, the translation module (120) can translate the registered post, and the post management module (110) can publish the translated post. (Step 1020)

[0100] The post management module (110) can transmit registered post data to a plurality of second computing devices (C2), respectively. The translation module (120) running on the plurality of second computing devices (C2) can input the received post data into the first generative machine learning model included in the second computing devices, thereby obtaining post data translated into different languages. The translation module (120) can store the translated post data as a file in the memory of the second computing device (C2) or add it as a record to a table in a database.

[0101] The post management module (110) running on the second computing device (C2) can publish translated posts so that the second user can view them.

[0102] Next, the post summary module (130) summarizes the registered posts, and the post management module (110) can publish the summarized posts. (Step 1030)

[0103] The post summary module (130) can input registered post data into a second generative machine learning model included in the first computing device (C1) to generate summarized text of the registered post.

[0104] The post summary module (130) transmits the summarized text to a plurality of second computing devices (C2), and the translation module (120) running on the plurality of second computing devices (C2) inputs the received summary text into a first generative machine learning model included in the second computing devices (C2), thereby generating summary text translated into different languages.

[0105] The post management module (110) running on the second computing device (C2) can issue a translated summary text for the second user to view.

[0106] Next, the keyword extraction module (140) extracts keywords from registered posts, and the post management module (110) can publish the extracted keywords. (Step 1040)

[0107] The keyword extraction module (140) can input registered post data into the third generative machine learning model included in the first computing device (C1) to extract keywords of the registered post.

[0108] The keyword extraction module (140) transmits the extracted keywords to each of a plurality of second computing devices (C2), and the translation module (120) running on the plurality of second computing devices (C2) inputs the received extracted keywords into the first generative machine learning model included in the second computing devices (C2), thereby obtaining extracted keywords translated into different languages.

[0109] The post management module (110) running on the second computing device (C2) can issue translated extracted keywords so that the second user can view them.

[0110] Next, the category selection module (150) selects a category for the registered post, and the post management module (110) can set the selected category. (Step 1050)

[0111] The category selection module (150) can input registered post data and category list data into the fourth generative machine learning model included in the first computing device (C1) to select a category of the registered post.

[0112] The post management module (110) can set the selected category as the category of the registered post so that a second user can view it.

[0113] Next, the quality measurement module (160) calculates the score of the registered post, and the post management module (110) can determine whether the post should be displayed. (Step 1060)

[0114] The quality measurement module (160) can input the registered post data into the fifth generative machine learning model included in the first computing device (C1) to calculate the score of the registered post data.

[0115] The post management module (110) can set exposure so that a second user can view the registered post if the score of the registered post data is equal to or greater than a reference score, and can set exposure prohibition so that a second user cannot view the registered post if the score is less than the reference score.

[0116] And the following steps (1030 to 1040) of the post translation step (1020) can be executed independently.

[0117] FIG. 10 is a diagram schematically showing the configuration of a computing device in one embodiment of the present invention.

[0118] The computing device (10) includes a processor (11), a memory device (12), an input / output device (13), and a system board (14).

[0119] The processor (11) executes an operation to read, change, or generate data used in one embodiment of the present invention. In addition, the processor (11) interprets and processes computer-readable instructions that execute a method of one embodiment of the present invention. The processor (11) may be a microprocessor including a control device that generates a control signal for interpreting and executing instructions, an arithmetic and logic operation device that executes arithmetic and logic operation instructions, a register that stores a plurality of instructions and the locations of the next instruction to be executed, input / output data, a cache memory that temporarily stores data exchanged between the processor (11) and a memory device (12), and a system bus that is a passage through which data moves within the processor (11).

[0120] The memory device (12) stores data processed or input / output within the computing device (10). In addition, the memory device (12) stores computer-readable instructions that execute a method according to an embodiment of the present invention. The memory device (12) may include a main memory device and an auxiliary memory device. The main memory device may include a random access memory device or a flash memory device. The auxiliary memory device may include one or more of a hard disk drive, a solid state drive (SSD), a flash memory device, an optical disc drive, and a magnetic tape.

[0121] The input / output device (13) inputs data into the computing device (10) and outputs data to the outside. In addition, the input / output device (13) inputs a computer-readable command that executes a method according to an embodiment of the present invention. The input / output device may include an external input / output terminal and a driver device that processes the external input / output terminal. For example, the external input / output terminal may include one or more of a serial port, a parallel port, a small computer system interface (SCSI), a universal serial bus (USB), IEEE 1394, an external serial advanced technology attachment (e-SATA), and Thunderbolt. In addition, the input / output device may include a network interface controller, and the network interface controller may be connected to a local area network (LAN) based on Ethernet in a wired manner, or to a wireless local area network (WLAN) based on Wi-Fi in a wireless manner.

[0122] The system board (14) connects between the processor (11), the memory device (12), and the input / output device (13), and provides a path for data processed by the computing device (10). The system board (14) may include an address bus, a command bus, a data bus, a chipset device that controls the bus, and a power system that supplies power to the components of the computer device.

[0123] The first and second computing devices (C1, C2) inside the system (100) and the external terminal devices (A, B) can be connected to a personal area network (PAN), a local area network (LAN), a metropolitan area network (MAN), or a wide area network (WAN), and data can be transmitted or received according to a data communication protocol such as TCP / IP (transmission control protocol / internet protocol), SMB (server message block), CIFS (common internet file system), or NFS (network file system).

[0124] Although the embodiments of the present invention have been described with reference to the attached drawings, those skilled in the art will appreciate that the present invention can be implemented in other specific forms without changing the technical concept or essential characteristics thereof. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. The scope of protection of the present invention should be interpreted by the following claims, and all technical ideas within the scope equivalent thereto should be interpreted as being included within the scope of the technical ideas defined by the present invention.

Claims

1. A method for translating summary text of a blog based on a generative machine learning model. A step in which a first computing device inputs post data into a generative machine learning model to generate summarized text; A step of transmitting summarized text by a first computing device to a plurality of second computing devices; and A step of having multiple second computing devices each input a summarized text into a generative machine learning model to generate summarized text translated into different languages; A method for automatically translating blog text based on a generative machine learning model.

2. In claim 1, A method of transmitting a translated summary text to a terminal device of a second user, wherein the plurality of second computing devices receive a message requesting to view the summarized text from the terminal device of the second user; A method for automatically translating blog text based on a generative machine learning model.

3. In claim 1, The post data received from the terminal device of the first user above includes text of the title, body, and subtitle. A method for automatically translating blog text based on a generative machine learning model.

4. In claim 1, A step in which a first computing device inputs post data into a generative machine learning model to select a category; and further comprising a step of setting the selected category as a category of the post data by the first computing device; A method for automatically translating blog text based on a generative machine learning model.

5. In claim 1, A step in which a first computing device inputs post data into a generative machine learning model to calculate a score; and The first computing device further includes a step of exposing the post data if the score of the post data is greater than or equal to a reference score, and not exposing the post data if the score is less than or equal to the reference score; A method for automatically translating blog text based on a generative machine learning model.

Images