Search

KR-20260065137-A - METHOD FOR AUTOMATICALLY GENERATING DOCUMENTS BASED ON ARTIFICIAL INTELLIGENCE MODEL AND DEVICE THEREOF

KR20260065137AKR 20260065137 AKR20260065137 AKR 20260065137AKR-20260065137-A

Abstract

The present invention relates to a method for automatically generating a document, comprising: receiving user data from a user terminal to obtain a first command; inputting the first command into a Large Language Model (LM) module to obtain a first response including a layout; inputting a second command based on the first response into the LLM module to obtain a second response including at least one frame, which is a sub-concept of the layout; inputting a third command based on the second response into the LLM module to obtain a third response including at least one block, which is a sub-concept of the frame; and generating a document based on the third response.

Inventors

  • 김성우

Assignees

  • 주식회사 사페레아우데

Dates

Publication Date
20260508
Application Date
20241101

Claims (12)

  1. In a method for automatically creating documents performed by at least one processor, A step of obtaining a first command by receiving user data from a user terminal; A step of inputting the above first command into a Large Language Model (LM) module to obtain a first response including a layout; A step of inputting a second command based on the first response into the LLM module to obtain a second response including at least one frame, which is a sub-concept of the layout; A step of inputting a third command based on the second response into the LLM module to obtain a third response including at least one block which is a sub-concept of the frame; and A step of generating a document based on the above third response Automatic document creation method.
  2. In Article 1, The above first command is obtained by the step of recognizing named entities in user data or the step of analyzing morphemes. Automatic document creation method.
  3. In Article 1, The step of obtaining the first response above A step of obtaining external server data based on the first command; A step of extracting keywords from the above external server data to extract a topic, which is a set of keywords; and A step comprising calculating the probability that the above keyword is formed among the above topics Automatic document creation method.
  4. In Article 1, The step of obtaining the first response above A step of obtaining external server data based on the first command; A step of extracting keywords from the external server data and calculating the similarity between the keywords and the external server data; and A step comprising repeating the step of extracting keywords from the external server data, excluding the keyword with the greatest similarity to the external server data. Automatic document creation method.
  5. In Article 1, The step of obtaining the second response above A step comprising generating nodes corresponding to the layout and the frame based on a graph neural network Automatic document creation method.
  6. In Article 5, The step of generating the second response above The method includes the step of creating a first node for a first layout and a second node for a first frame, and The step of generating the above document is, A step of determining context similarity for the first node and the second node; and If the above context similarity is below a reference value, the method includes the step of modifying the first layout or the first frame. Automatic document creation method.
  7. In Article 5, The step of generating the second response above The method includes the step of creating a first node for a first layout, a second node for a first frame, and a third node for a second frame, and The step of generating the above document is, A step of determining context similarity for the second node and the third node; and If the above context similarity is below a reference value, the method includes the step of modifying the first layout, the first frame, or the second frame. Automatic document creation method.
  8. In claim 6 or 7, The above modification step A step of aggregating feature vectors of the above nodes; A step comprising multiplying the above feature vector by a weight Automatic document creation method.
  9. In Article 1, A step of obtaining external server data based on the first command above; A further step of constructing a database by storing the aforementioned acquired external server data Automatic document creation method.
  10. In Article 9, The above third response is generated by the above database, and A step of generating a fourth command to determine the reliability of the third response; A step of determining the reliability of the third response based on the fourth response, which is a response to the fourth command; and If deemed reliable, the step of generating a document based on the third response further includes Automatic document creation method.
  11. A computer program stored on a computer-readable recording medium to execute any one of the document automatic creation methods of paragraphs 1 through 10.
  12. Communication module; memory for storing instructions; and It includes at least one processor connected to the memory and configured to execute at least one computer-readable program included in the memory, and The above at least one program is, Obtaining user data from a user terminal and obtaining a first command, Input the above first command into the LLM (Large Language Model) module to obtain a first response including a layout containing at least one table of contents, and A second command based on the first response is input into the LLM module to obtain a second response including at least one frame, which is a sub-concept of the layout, and A third command based on the second response is input into the LLM module to obtain a third response including at least one block, which is a sub-concept of the frame, and Generating a document based on the above third response Automatic document creation device.

Description

Method for automatically generating documents based on an artificial intelligence model and device therefor The present invention relates to an artificial intelligence model-based automatic document creation method and apparatus thereof, and more specifically, to a method and apparatus for creating a document that handles complex topics and maintains consistency. Existing document generation systems primarily operated by generating fragmentary text based on user input or filling in predefined templates. This approach makes it difficult to maintain the logical flow and consistency of the entire document and has limitations in handling high-level content or complex topics. In addition, there were issues where it was limited to simply generating sentences or paragraphs based on user input, lacked flexibility, and made it difficult to generate customized documents that met user requirements. Accordingly, the present invention aims to provide a method for generating a document in which the logic of the document is consistent and a personalized document is generated based on user input data. FIG. 1 is a drawing showing a document automatic creation service system according to one embodiment. FIG. 2 is a diagram showing the configuration of a computing device according to one embodiment. FIGS. 3 and FIGS. 4 are diagrams showing the steps of sequentially acquiring commands and automatically generating a document according to one embodiment. FIG. 5 is a figure showing the steps of a method for obtaining a first response, including the step of extracting a topic by extracting keywords according to one embodiment. FIG. 6 is a figure showing the steps of a method for obtaining a first response, including the step of extracting the most similar keyword according to one embodiment. FIG. 7 is a diagram illustrating a graph neural network having a layout, a frame, and a block as nodes according to one embodiment. FIG. 8 is a block diagram of a computing device that automatically creates a document according to one embodiment. Hereinafter, exemplary embodiments according to the present invention will be described in detail with reference to the contents described in the attached drawings. However, the present invention is not limited or restricted by exemplary embodiments. Unless otherwise defined, all terms used in this specification (including technical and scientific terms) shall be used in a meaning that is commonly understood by those skilled in the art to which this disclosure belongs, but this may vary depending on the intent of those skilled in the art, case law, the emergence of new technology, etc. Furthermore, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise. In certain cases, terms have been selected at the applicant's discretion, and in such cases, their meanings will be described in detail in the relevant explanatory sections. Accordingly, terms used in this disclosure should be defined not merely by their names, but based on their meanings and the content throughout this disclosure. Throughout this specification, when a part is described as "comprising" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components. Furthermore, the singular form used in this specification includes the plural form unless specifically stated otherwise. Additionally, the expression "at least one of a, b, and/or c" as used throughout this specification may encompass 'a alone', 'b alone', 'c alone', 'a and b', 'a and c', 'b and c', or 'a, b, and c all'. Meanwhile, terms such as "first and/or second" used in this specification may be used to describe various components, but they are used solely for the purpose of distinguishing one component from another and are not intended to limit the scope to the components referred to by such terms. For example, without departing from the scope of the present invention, the first component may be named the second component, and the second component may also be named the first component. Additionally, terms such as “…part,” “…module,” etc., as described in this specification refer to a unit that processes at least one function or operation, which may be implemented in hardware or software, or a combination of hardware and software. Furthermore, embodiments of this disclosure may be represented in this specification by functional block configurations and various processing steps. These functional blocks may be implemented by various numbers of hardware and/or software configurations that execute specific functions. For example, embodiments of this disclosure may employ integrated circuit configurations such as memory, processing, logic, look-up tables, etc., which can execute various functions under the control of one or more microprocessors or other control devices. In an embodiment according to the present disc