Apparatus and method for modifying content

Abstract

Disclosed are an apparatus and a method for modifying content. An apparatus for modifying content according to one aspect of the present invention comprises: a communication module; a memory; and a processor connected to the communication module and the memory. The processor may analyze the structure of content to identify a structure item, generate modification information on at least one of keywords or the substance of the content, select a template having the highest similarity to the structure item of the content among pre-stored templates, compare the selected template and the structure item to generate additional supplementary information, transmit modification notification information including at least one of the modification information or the additional supplementary information to a user terminal through the communication module, and when a modification signal is received from the user terminal, modify the content by applying the selected template and at least one of the modification information or the additional supplementary information.

Description

Content modification device and method

[0001] The present invention relates to a content correction device and method that can automatically correct at least one of typos, content, and templates included in content.

[0002]

[0003] Thanks to advances in information and communication technology, consumers now possess not only desktop computers but also various mobile devices, such as smartphones and tablets, and use them to consume a variety of content, including movies, images, and news articles. This content can be for entertainment or business purposes, but it also includes educational content for general learning purposes.

[0004] However, content such as learning content may contain typos and unverified information, which is inconvenient as users have to find and correct these themselves.

[0005] The background technology of the present invention is disclosed in Korean Patent Publication No. 10-2013-0089998 (August 13, 2013), entitled ‘Learning information provision system and method for providing learning information to learners using the same.’

[0006]

[0007] The present invention has been devised to improve the above-mentioned problems, and an object of the present invention is to provide a content correction device and method that can automatically correct at least one of typos, content, and templates included in content.

[0008] The problems to be solved by the present invention are not limited to the problem(s) mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

[0009]

[0010] According to one aspect of the present invention, a content modification device includes a communication module, a memory, and a processor connected to the communication module and the memory, wherein the processor analyzes the structure of content to identify a structure item, generates modification information for at least one of a keyword and content in the content, selects a template having the highest similarity to a structure item of the content among pre-stored templates, compares the selected template with the structure item to generate additional supplementary information, and transmits modification notification information including at least one of the modification information and the additional supplementary information to a user terminal through the communication module, and when a modification signal is received from the user terminal, the content can be modified by applying at least one of the selected template, the modification information, and the additional supplementary information.

[0011] In the present invention, the processor can analyze the structure of the content based on multimodal generative AI.

[0012] In the present invention, the processor can identify structural items of the content by extracting keywords from the content and performing semantic connectivity and semantic analysis between sentences or paragraphs based on the extracted keywords.

[0013] In the present invention, the processor can extract keywords from the content, extract an error keyword from among the extracted keywords, obtain a replacement keyword for the error keyword, and generate typo correction information including the error keyword and the replacement keyword.

[0014] In the present invention, if there is an item requiring authenticity verification among the structural items of the content, the processor can generate authenticity verification request information including the contents of the corresponding structural item, obtain the latest information about the contents of the corresponding structural item included in the authenticity verification request information using a multimodal generative AI, and generate content modification information including the contents and latest information of the corresponding structural item.

[0015] In the present invention, the processor generates query information including a field, summary, and content of a configuration item among the configuration items of the template that does not exist in the structure item of the content, applies the query information to a multimodal generative AI to obtain response information for the corresponding configuration item, and generates the obtained response information as the additional supplementary information.

[0016]

[0017] A content modification method according to another aspect of the present invention may include a step in which a processor analyzes a structure of content to identify a structure item, a step in which the processor generates modification information for at least one of a keyword and content in the content, a step in which the processor selects a template having the highest similarity to a structure item of the content from among pre-stored templates and compares the selected template with the structure item to generate additional supplementary information, a step in which the processor transmits modification notification information including at least one of the modification information and the additional supplementary information to a user terminal, and a step in which the processor modifies the content by applying at least one of the selected template, the modification information, and the additional supplementary information when receiving a modification signal from the user terminal.

[0018] In the present invention, in the identifying step, the processor can analyze the structure of the content based on multimodal generative AI.

[0019] In the present invention, in the identifying step, the processor extracts keywords from the content and performs semantic connectivity and semantic analysis between sentences or paragraphs based on the extracted keywords, thereby identifying structural items of the content.

[0020] In the step of generating the correction information of the present invention, the processor can extract keywords from the content, extract an error keyword from among the extracted keywords, obtain a replacement keyword for the error keyword, and generate typo correction information including the error keyword and the replacement keyword.

[0021] In the present invention, in the step of generating the above modification information, if there is an item among the structural items of the content that requires verification of authenticity, the processor generates authenticity verification request information including the contents of the corresponding structural item, obtains the latest information on the contents of the corresponding structural item included in the authenticity verification request information using a multimodal generation type AI, and generates content modification information including the contents and latest information of the corresponding structural item.

[0022] In the present invention, in the step of generating the additional supplementary information, the processor generates query information including a field, summary, and content of a configuration item among the configuration items of the template that does not exist in the structure item of the content, applies the query information to a multimodal generative AI to obtain response information for the corresponding configuration item, and generates the obtained response information as the additional supplementary information.

[0023] In addition, other methods for implementing the present invention, other systems and computer programs for executing the above methods may be further provided.

[0024]

[0025] A content modification device and method according to one embodiment of the present invention can automatically modify at least one of a typo, content, and template included in content.

[0026] Meanwhile, the effects of the present invention are not limited to the effects mentioned above, and various effects may be included within a range obvious to those skilled in the art from the contents described below.

[0027]

[0028] FIG. 1 is a drawing for explaining a content modification system according to one embodiment of the present invention.

[0029] FIG. 2 is a block diagram schematically illustrating a content modification device according to one embodiment of the present invention.

[0030] FIG. 3 is a flowchart illustrating a content modification method according to one embodiment of the present invention.

[0031]

[0032] Hereinafter, a content modification device and method according to one embodiment of the present invention will be described with reference to the attached drawings. In this process, the thickness of lines and the sizes of components depicted in the drawings may be exaggerated for clarity and convenience.

[0033] Furthermore, the implementations described herein may be implemented as, for example, a method or process, an apparatus, a software program, a data stream, or a signal. Even if discussed only in the context of a single form of implementation (e.g., discussed only as a method), the implementation of the discussed features may also be implemented in other forms (e.g., as an apparatus or a program). The apparatus may be implemented using suitable hardware, software, firmware, and the like. The method may be implemented in an apparatus such as a processor, which generally refers to a processing device including, for example, a computer, a microprocessor, an integrated circuit, or a programmable logic device. A processor also includes a communication device such as a computer, a cell phone, a personal digital assistant ("PDA"), and other devices that facilitate communication of information between end-users.

[0034] In addition, the terminology used in this specification is only used to describe specific embodiments and is not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, it should be understood that the terms "comprise" or "have" are intended to specify the presence of a feature, number, step, operation, component, part, or combination thereof described in the specification, but do not exclude in advance the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof. Terms such as first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another.

[0035] Hereinafter, embodiments according to the present invention will be described in detail with reference to the attached drawings. In describing with reference to the attached drawings, identical or corresponding components are assigned the same drawing numbers, and redundant descriptions thereof will be omitted.

[0036]

[0037] FIG. 1 is a drawing for explaining a content modification system according to one embodiment of the present invention.

[0038] Referring to FIG. 1, a content modification system according to one embodiment of the present invention includes a user terminal (100) and a content modification device (200).

[0039] The user terminal (100) can execute and / or display a content modification application or content modification site provided by the content modification device (200), and the content modification device (200) that receives the user's access identification information (ID) and password through the user terminal (100) can perform user authentication for the content modification application or content modification site.

[0040] A user terminal (100) can request content modification from a content modification device (200) through a content modification application or a content modification site, and receive modified content from the content modification device (200). Here, the content can be diverse, such as learning content, lecture content, movies, images, and news articles.

[0041] The user terminal (100) may receive modification notification information including modification information and additional supplementary information from the content modification device (200), and may receive a modification signal including an automatic modification command or a selective modification command from the user and transmit the signal to the content modification device. The automatic modification command may be a command that allows modification of all information included in the modification notification information, and the selective modification command may be a command that allows modification of only information selected by the user from among the information included in the modification notification information.

[0042] The user terminal (100) may be, but is not limited to, a desktop computer, a smartphone, a laptop, a tablet PC, a smart TV, a mobile phone, a PDA (personal digital assistant), a laptop, a media player, a micro server, a GPS (global positioning system) device, an e-book reader, a digital broadcasting terminal, a navigation device, a kiosk, an MP3 player, a digital camera, a home appliance, or any other mobile or non-mobile computing device operated by a user. In addition, the user terminal (100) may be a wearable terminal such as a watch, glasses, a hair band, or a ring, having a communication function and a data processing function. The user terminal (100) is not limited to the above-described contents, and any terminal capable of web browsing may be borrowed without limitation.

[0043] When a content modification request signal is received from a user terminal (100), the content modification device (200) analyzes the structure of the content to identify structural items, generates modification information that requires modification in the content, selects a template with the highest similarity to the structural items of the content among the previously stored templates, and compares the selected template with the structural items to generate additional supplementary information.

[0044] The content modification device (200) transmits modification notification information including modification information and additional supplementary information to the user terminal (100), and when receiving a modification signal from the user terminal (100), the content can be automatically modified by applying at least one of the selected template, modification information, and additional supplementary information.

[0045] At this time, the content modification device (200) can modify the content using multi-modal generative AI.

[0046] Multimodal AI can learn from a variety of modalities, including text, images, voice, audio, video, 3D, LiDAR, and radio frequency (RF) sensor data. Multimodal generative AI can use all these sensors to fuse data, gaining a more holistic understanding of the world and providing improved answers based on that understanding.

[0047] The content modification device (200) may be implemented using one or more servers, but the present invention is not necessarily limited thereto, and may also be implemented using various devices capable of information processing, such as a dedicated device or a cloud system in which multiple information processing devices are linked.

[0048] For a detailed description of this content modification device (200), refer to FIG. 2.

[0049]

[0050] FIG. 2 is a block diagram schematically illustrating a content modification device according to one embodiment of the present invention.

[0051] Referring to FIG. 2, a content modification device (200) according to one embodiment of the present invention includes a communication module (210), a memory (220), a database (230), and a processor (240).

[0052] The communication module (210) may provide a communication interface necessary to provide transmission and reception signals in the form of packet data between the content modification device (200) and the user terminal (100) in conjunction with a communication network. Furthermore, the communication module (210) may serve to receive a content modification request signal including content from the user terminal (100) and may serve to transmit the content modified by the processor (240) to the user terminal (100). Here, the communication network refers to a medium that serves to connect the content modification device (200) and the user terminal (100), and may include a path that provides a connection path so that the user terminal (100) can transmit and receive information after connecting to the content modification device (200). In addition, the communication module (210) may be a device including hardware and software necessary to transmit and receive signals such as control signals or data signals through a wired or wireless connection with another network device. In addition, the communication module (210) can be implemented in various forms such as a short-range communication module, a wireless communication module, a mobile communication module, and a wired communication module.

[0053] The memory (220) is a configuration that stores data related to the operation of the content modification device (200). Here, the memory (220) may use a known storage medium, and for example, may use any one or more of known storage media such as ROM, PROM, EPROM, E EPROM, RAM, etc. In particular, the memory (220) may store a program (application or applet) that analyzes the structure of the content, generates modification information that needs to be modified in the content, compares the structure items of the content with a selected template to generate additional supplementary information, and modifies the content by applying at least one of the selected template, the modification information, and the additional supplementary information, and the stored information may be selected by the processor (240) as needed.

[0054] The database (230) may include a user database (232), a content database (234), and a template database (236).

[0055] The user database (232) can store user information of each user, content created or modified by each user, etc. Here, the user information can include basic information about the user, such as the user's name, affiliation, personal information, gender, age, contact information, email, address, occupation, title, job, and competency level, as well as information related to authentication (login) such as ID (or email) and password, country of access, location of access, information about the device used for access, and network environment of access.

[0056] The content database (234) can store content by category. Each content stored in the content database (234) is configured with information such as content complexity, learning type, learning difficulty, and applied template. Additionally, each content stored in the content database (234) is registered with a summary, keywords, and content.

[0057] The template database (236) can store template information including template identification information, configuration items of each template, application conditions of each template, etc. Here, the configuration items of the template include content introduction, motivational content, content required for learning, quizzes, summaries, demonstration content, missions for practice, simulation content, concept explanation of learning content, problem raising content, hypothesis proposition content, search tool and opinion collection content (configured to provide a search tool so that opinions can be input), presentation of discussion space (url) online / offline, items that allow checking the progress after discussion, content that allows learning by creating a scenario, content that allows creating it, evaluation items, items that allow checking the learning content, etc., and the configuration items of the template can be set differently depending on the type of template.

[0058] The template application conditions include simple, complex, understanding-centered, practice-centered, and learning difficulty levels, and may vary depending on the type of template. For example, the case where there are templates 1 through 8 will be described. In this case, the application conditions of the first template may be understanding-centered and learning difficulty level 1 or 2, and the application conditions of the second template may be simple, understanding-centered, and learning difficulty level 3. In addition, the application conditions of the third template may be simple, practice-centered, and learning difficulty level 1, or complex, practice-centered, and learning difficulty level 1. The application conditions of the fourth template may be simple, practice-centered, and learning difficulty level 2 or 3. The application conditions of the fifth template may be complex, understanding-centered, and learning difficulty level 1, and the application conditions of the sixth template may be complex, understanding-centered, and learning difficulty level 2 or 3. The application conditions of the 7th template may be complex, practice-oriented, and learning difficulty level 2, and the application conditions of the 8th template may be complex, practice-oriented, and learning difficulty level 3.

[0059] Meanwhile, in the embodiment of the present invention, the database (230) is described as being included in the content modification device (200), but the database (230) may be provided in an external device connected to the content modification device (200) via a wired or wireless communication network.

[0060] The processor (240) may be configured to control the overall operation of the content modification device (200). For example, the processor (240) may execute software (e.g., a program) stored in the memory (220) to control components connected to the processor (240) (e.g., at least one of the communication module (210), the memory (220), and the database (230). The processor (240) may be implemented as, but is not limited to, an Application Specific Integrated Circuit (ASIC), a Digital Signal Processor (DSP), a Programmable Logic Devices (PLD), a Field Programmable Gate Arrays (FPGAs), a Central Processing unit (CPU), a microcontroller, and / or a microprocessor.

[0061] The processor (240) analyzes the structure of content to identify a structure item, generates modification information for at least one of a keyword and content in the content, selects a template having the highest similarity to a structure item of the content among pre-stored templates, compares the selected template with the structure item of the content to generate additional supplementary information, and transmits modification notification information including at least one of the modification information and the additional supplementary information to the user terminal (100) through the communication module (210), and when receiving a modification signal from the user terminal (100), the content can be modified by applying at least one of the selected template, the modification information, and the additional supplementary information.

[0062] Hereinafter, the operation of the processor (240) will be described in detail.

[0063] When a content modification request signal including content is received from a user terminal (100), the processor (240) can analyze the structure of the content to identify structure items. Here, the content may be various, such as content stored in a content database (234), user-generated content, etc.

[0064] The processor (240) can analyze the structure of content based on multimodal generative AI.

[0065] Additionally, the processor (240) can identify structural items of the content by extracting keywords from the content and performing semantic connectivity and semantic analysis between sentences or paragraphs based on the extracted keywords.

[0066] In addition, the processor (240) may analyze the structure of the content using a document structure analysis model to identify structural items. Here, the document structure analysis model may be a model for generating structural information about an input document image when a document image is input. The document structure analysis model may be, for example, a model based on a deep neural network such as a mask regional convolutional neural network (MRCNN), but may also be one of various known document structure analysis models that can be learned through various machine learning techniques.

[0067] In addition, the processor (240) can extract morphemes, which are the minimum semantic units of a word, from content, and structure the dependency relationship of morphemes to understand the structure and meaning of characters, and can also identify structural items of the content based on this.

[0068] For example, the processor (240) can analyze the content to identify structural items such as what is described in the section, what is an example in the case of a section that explains with a definition, content description, example description, example (study case, etc.), and a section that requires motivation, hypothesis, or demonstration.

[0069] The processor (240) may generate correction information that requires correction in the content. Here, the correction information may include typo correction information including error keywords and replacement keywords, and content correction information including the content and latest information of the structure item.

[0070] The processor (240) can recognize keywords in content, extract erroneous keywords from among the recognized keywords, obtain alternative keywords for the erroneous keywords, and generate typo correction information including the erroneous keywords and alternative keywords.

[0071] Specifically, the processor (240) can recognize text from image-type content using optical character recognition (OCR) and extract keywords from the recognized text. At this time, the processor (240) can recognize text and extract keywords using a deep learning-based OCR model such as a convolutional neural network (CNN), a recurrent neural network (RNN), or a CRNN method that combines CNN and RNN.

[0072] Once keywords are extracted, the processor (240) can compare the extracted keywords with keywords stored in a previously stored keyword dictionary DB (not shown) to extract erroneous keywords including typos, etc. Once erroneous keywords are extracted, the processor (240) can obtain replacement keywords to correct the erroneous keywords from the keyword dictionary DB and generate typo correction information including the erroneous keywords and replacement keywords. Here, the replacement keywords can be normal keywords that replace the erroneous keywords.

[0073] When there is an item requiring authenticity verification among the structural items of the content, the processor (240) can generate authenticity verification request information including the contents of the corresponding structural item, obtain the latest information about the contents of the corresponding structural item included in the authenticity verification request information using a multimodal generative AI, and generate content modification information including the contents and latest information of the corresponding structural item.

[0074] That is, the processor (240) can determine whether there are items requiring fact verification, case verification, etc., among the structural items of the content. For example, in the case of structural items including keywords such as 'recent~', 'fact~', 'for example~', etc., it can be determined whether there are items requiring fact verification, case verification, etc.

[0075] If there are items requiring fact verification, case verification, etc., the processor (240) may determine that there are items requiring authenticity verification among the structural items of the content. If there are items requiring authenticity verification, the processor (240) may generate authenticity verification request information including the contents of the corresponding structural items, and may obtain the latest information about the contents of the corresponding structural items included in the authenticity verification request information using multimodal generative AI. The processor (240) may compare the contents of the corresponding structural items with the latest information, and if the contents differ from the latest information, the processor may determine that the contents are not true. If the contents are not true, the processor (240) may generate content modification information including the contents of the corresponding structural items and the latest information.

[0076] The processor (240) can select a template with the highest similarity to the structural item of the content among the previously stored templates, and compare the selected template with the structural item to generate additional supplementary information.

[0077] That is, the processor (240) can compare the structural items of the content with the templates stored in the template database (236) to select the most similar template. For example, the processor (240) can compare the structural items of the content with the configuration items of the templates stored in the template database (236) to select the template that includes the largest number of structural items of the content.

[0078] When the template most similar to the structural items of the content is selected, the processor (240) can determine whether there is a configuration item among the configuration items of the selected template that does not exist in the structural items of the content.

[0079] For a configuration item that exists in a structure item of content among the configuration items of the selected template, the processor (240) can apply (input, insert) the contents of the structure item of content to the configuration item of the template.

[0080] In the case of a configuration item that does not exist in the content structure item among the configuration items of the selected template, the processor (240) generates query information including the field and the configuration item content, applies the query information to a multimodal generative AI (artificial intelligence model) to obtain response information for the corresponding configuration item, and can generate the obtained response information as additional supplementary information. Here, the additional supplementary information can include response information to be applied (input, inserted) to a configuration item that does not exist in the content structure item among the configuration items of the selected template.

[0081] The processor (240) can transmit modification notification information including modification information and additional supplementary information to the user terminal (100) through the communication module (210).

[0082] The user terminal (100) can display modification notification information. At this time, the user terminal (100) can display a modification notification screen including modification information, additional supplementary information, automatic modification commands, and selective modification commands. The user can check the modification information and additional supplementary information, and select the automatic modification command or the selective modification command. If the user selects the automatic modification command, the processor (240) can automatically modify both the modification information and the additional supplementary information. If the user selects the selective modification command, the processor (240) can automatically modify only the items (contents) selected by the user.

[0083] When a modification signal is received from the user terminal (100), the processor (240) can modify the content by applying at least one of the selected template, modification information, and additional supplementary information.

[0084] For example, when a correction signal including a correction command for typo correction information is received, the processor (240) may apply the contents of the structure item of the content to the contents of the configuration item of the corresponding template, and correct the error keyword with a replacement keyword. At this time, the contents of the configuration item that does not exist in the content structure item among the configuration items of the template may be empty.

[0085] When a modification signal containing only a modification command for content modification information is received, the processor (240) applies the content of the content structure item to the content of the corresponding template's configuration item, but may modify the content of the content structure item to the latest information. For example, an image included in the content may be modified to an image reflecting the latest trends. At this time, the content of a configuration item that does not exist in the content structure item among the template's configuration items may be empty.

[0086] When a modification signal including a modification command for additional supplementary information is received, the processor (240) applies the contents of the structure item of the content to the contents of the configuration item of the corresponding template, but may apply the additional supplementary information to a configuration item that does not exist in the content structure item among the configuration items of the template.

[0087] When a correction signal including an automatic correction command is received, the processor (240) may apply the contents of the structure item of the content to the contents of the configuration item of the corresponding template, correct an error keyword to a replacement keyword, correct the contents of the content structure item to the latest information, and apply additional supplementary information to a configuration item that does not exist in the content structure item among the configuration items of the template.

[0088] Once the content modification is complete, the processor (240) can transmit the modified content to the user terminal (100). At this time, the processor (240) can store the modified content in the user database (232) or the content database (234).

[0089]

[0090] FIG. 3 is a flowchart illustrating a content modification method according to one embodiment of the present invention.

[0091] Referring to FIG. 3, when a content modification request signal is received from a user terminal (100) (S302), the processor (240) analyzes the structure of the content to identify structure items (S304).

[0092] At this time, the processor (240) can analyze the structure of the content based on multimodal generative AI. In addition, the processor (240) can identify the structural items of the content by extracting keywords from the content and performing semantic connectivity and meaning analysis between sentences or paragraphs based on the extracted keywords. In addition, the processor (240) can analyze the structure of the content using a document structure analysis model to identify the structural items. In addition, the processor (240) can extract morphemes, which are the minimum semantic units of words, from the content and structure the dependency relationships of the morphemes to understand the structure and meaning of characters, and can identify the structural items of the content based on this.

[0093] When step S304 is performed, the processor (240) generates typo correction information including error keywords and replacement keywords in the content (S306).

[0094] The processor (240) can perform text recognition and keyword extraction using a deep learning-based OCR model, such as a convolutional neural network (CNN), a recurrent neural network (RNN), or a CRNN method combining CNN and RNN. Then, the processor (240) can compare the extracted keywords with keywords stored in a keyword dictionary database to extract error keywords, including typos. Then, the processor (240) can obtain alternative keywords for the error keywords and generate error correction information, including the error keywords and alternative keywords.

[0095] When step S306 is performed, the processor (240) generates content modification information including the content and latest information of the structure item (S308).

[0096] That is, when there is an item requiring authenticity verification among the structural items of the content, the processor (240) can generate authenticity verification request information including the contents of the corresponding structural item, obtain the latest information about the contents of the corresponding structural item included in the authenticity verification request information using a multimodal generative AI, and generate content modification information including the contents and latest information of the corresponding structural item.

[0097] When step S308 is performed, the processor (240) selects a template having the highest similarity to the structural item of the content among the templates stored in the template database (236) (S310), and compares the selected template with the structural item to generate additional supplementary information (S312).

[0098] When the template most similar to the structural item of the content is selected, the processor (240) can determine whether there is a structural item among the structural items of the selected template that does not exist in the structural item of the content. In the case of a structural item among the structural items of the selected template that exists in the structural item of the content, the processor (240) can input the content of the structural item of the content into the structural item of the template. In the case of a structural item among the structural items of the selected template that does not exist in the structural item of the content, the processor (240) can generate query information including a field, a summary, and the content of the structural item, apply the query information to a multimodal generative AI (artificial intelligence model) to obtain response information for the corresponding structural item, and generate the obtained response information as additional supplementary information.

[0099] When step S312 is performed, the processor (240) transmits modification notification information including modification information and additional supplementary information to the user terminal (100) (S314). Here, the modification notification information may include modification information, additional supplementary information, an automatic modification command, a selective modification command, etc.

[0100] After performing step S314, if a modification signal is received from the user terminal (100) (S316), the processor (240) modifies the content by applying at least one of the selected template, modification information, and additional supplementary information (S318).

[0101] When a correction signal including a correction command for typo correction information is received, the processor (240) may apply the contents of the structure item of the content to the contents of the configuration item of the corresponding template, and correct the error keyword with a replacement keyword. At this time, the contents of the configuration item that does not exist in the content structure item among the configuration items of the template may be empty.

[0102] When a modification signal containing only a modification command for content modification information is received, the processor (240) applies the content of the content structure item to the content of the corresponding template's configuration item, but may modify the content of the content structure item to the latest information. For example, an image included in the content may be modified to an image reflecting the latest trends. At this time, the content of a configuration item that does not exist in the content structure item among the template's configuration items may be empty.

[0103] When a modification signal including a modification command for additional supplementary information is received, the processor (240) applies the contents of the structure item of the content to the contents of the configuration item of the corresponding template, but may apply the additional supplementary information to a configuration item that does not exist in the content structure item among the configuration items of the template.

[0104] When step S316 is performed, the modified content is transmitted to the user terminal (100) (S318). At this time, the processor (240) may store the modified content in the user database (232) or content database (234).

[0105]

[0106] As described above, the content modification device and method according to one embodiment of the present invention can automatically modify at least one of a typo, content, and template included in the content.

[0107] While the present invention has been described with reference to the embodiments illustrated in the drawings, these are merely exemplary, and those skilled in the art will understand that various modifications and equivalent embodiments are possible. Therefore, the true technical protection scope of the present invention should be defined by the following claims.

Claims

1. Communication module; memory; and Includes a processor connected to the above communication module and memory, The above processor, A content modification device characterized in that it analyzes the structure of content to identify a structure item, generates modification information for at least one of a keyword and content in the content, selects a template having the highest similarity to a structure item of the content among pre-stored templates, compares the selected template with the structure item to generate additional supplementary information, transmits modification notification information including at least one of the modification information and the additional supplementary information to a user terminal through the communication module, and modifies the content by applying at least one of the selected template, the modification information, and the additional supplementary information when a modification signal is received from the user terminal.

2. In paragraph 1, The above processor, A content modification device characterized by analyzing the structure of the content based on multimodal generative AI.

3. In paragraph 1, The above processor, A content modification device characterized in that it identifies structural items of the content by extracting keywords from the content and performing semantic connection and semantic analysis between sentences or paragraphs based on the extracted keywords.

4. In paragraph 1, The above processor, A content correction device characterized by extracting keywords from the content, extracting error keywords from the extracted keywords, obtaining alternative keywords for the error keywords, and generating typo correction information including the error keywords and alternative keywords.

5. In paragraph 1, The above processor, A content modification device characterized in that, if there is an item requiring authenticity verification among the structural items of the above content, authenticity verification request information including the contents of the corresponding structural item is generated, the latest information on the contents of the corresponding structural item included in the authenticity verification request information is obtained using a multimodal generative AI, and content modification information including the contents and latest information of the corresponding structural item is generated.

6. In paragraph 1, The above processor, A content modification device characterized in that, for a configuration item that does not exist in the structure item of the content among the configuration items of the above template, query information including a field, summary, and configuration item content is generated, the query information is applied to a multimodal generative AI to obtain response information for the corresponding configuration item, and the obtained response information is generated as the additional supplementary information.

7. A step in which the processor analyzes the structure of the content to identify structural items; The step of the processor generating modification information for at least one of a keyword and content in the content; A step in which the processor selects a template having the highest similarity to a structural item of the content among the previously stored templates, and compares the selected template with the structural item to generate additional supplementary information; The step of the processor transmitting, to the user terminal, modification notification information including at least one of the modification information and additional supplementary information; and A step in which the processor, upon receiving a modification signal from the user terminal, modifies the content by applying at least one of the selected template, modification information, and additional supplementary information; A content modification method characterized by including:

8. In paragraph 7, In the above identification step, A content modification method characterized in that the above processor analyzes the structure of the content based on multimodal generative AI.

9. In paragraph 7, In the above identification step, A content modification method characterized in that the processor extracts keywords from the content and performs semantic connectivity and semantic analysis between sentences or paragraphs based on the extracted keywords, thereby identifying structural items of the content.

10. In paragraph 7, In the step of generating the above modification information, A content correction method characterized in that the processor extracts keywords from the content, extracts an error keyword from the extracted keywords, obtains a replacement keyword for the error keyword, and generates typo correction information including the error keyword and the replacement keyword.

11. In paragraph 7, In the step of generating the above modification information, A content modification method characterized in that the processor, when there is an item requiring authenticity verification among the structural items of the content, generates authenticity verification request information including the contents of the corresponding structural item, obtains the latest information about the contents of the corresponding structural item included in the authenticity verification request information using a multimodal generative AI, and generates content modification information including the contents of the corresponding structural item and the latest information.

12. In paragraph 7, In the step of generating the above additional supplementary information, A content modification method characterized in that the processor generates query information including a field, summary, and content of a configuration item among the configuration items of the template that does not exist in the structure item of the content, applies the query information to a multimodal generative AI to obtain response information for the corresponding configuration item, and generates the obtained response information as the additional supplementary information.

Images