KR-20260066821-A - Software and Method for Providing a Book Writing Environment

Abstract

The present invention relates to software and a method for providing an environment for writing a book, with the aim of helping users write books more efficiently. To this end, the user can (a) configure a customized table of contents by selecting a topic and answering questions generated by an LLM, and (b) proceed with writing in paragraph units through an editor. The software enhances the efficiency of the writing process by providing (c) drafting, (d) automatic sentence recommendations, (e) similar word recommendations, and (f) revision functions. Additionally, it includes (g) a personalized writing environment based on custom setting information and (h) a function to export the written book in PDF or DOCX format.

Inventors

• 김준형
• 강돈혁
• 김영상

Assignees

• 김준형

Dates

Publication Date

20260512

Application Date

20241105

Claims (6)

1. As software that supports book writing, (a) A function that automatically generates questions using a Large Language Model (LLM) while the user inputs topics, and constructs a customized table of contents based on the user's responses to the questions; (b) A user experience (UX) feature that conveniently saves book customization information, such as style, writing intent, target audience, and personal style, within the writing process without user input; (c) A feature that creates a draft based on a configured table of contents and saved custom settings, and provides customized automatic sentence recommendations and proofreading functions; (d) Software characterized in that the functions of (a), (b), and (c) are organically connected and interact as a single software, producing synergistic effects to provide the user with a more efficient and effective writing environment.
2. In claim 1, Software that includes a function to automatically generate a draft suitable for the user's context by utilizing stored custom setting information without direct input from the user during the table of contents configuration process.
3. In claim 1, Software that includes an automatic sentence recommendation function that recommends the next sentence based on the previous sentence written by the user and custom setting information.
4. In claim 1, Software that includes a proofreading function that automatically reviews grammatical errors and contextual problems in text written by a user and provides correction suggestions.
5. In claim 1, Software characterized by providing a personalized writing environment to the user based on saved book customization information.
6. In claim 1, Software characterized by maximizing the overall efficiency of book writing by integrating the above functions to provide synergistic effects to the user.

Description

Software and Method for Providing a Book Writing Environment Software and Method for Providing a Book Writing Environment The technical field of this software relates to artificial intelligence-based content creation and editing software. Specifically, it concerns software that supports users in efficiently writing books by providing a book writing environment utilizing Large Language Models (LLM) and focusing on automated text generation and editing functions. This invention belongs to fields such as artificial intelligence natural language processing, document editing tools, user interfaces (UI/UX), and cloud-based data storage and processing, and in particular corresponds to application fields such as automatic document generation, drafting assistance systems, and personalized text recommendation technologies. In addition, this invention also serves as a digital publishing tool that runs on various platforms, including web applications and mobile applications. Large Language Models (LLMs) are a type of artificial intelligence trained on large text datasets to generate human-like responses to natural language inputs; they are language models composed of artificial neural networks possessing numerous parameters (typically billions of weights or more). These LLMs can be trained on significant amounts of unlabeled text using self-supervised or semi-self-supervised learning. As the performance of LLM improves rapidly, the need to automate business operations within companies using LLM is increasing; however, currently, there is no way to actually implement the automation of various tasks quickly using LLM. Figure 1 is a configuration diagram of book writing software. Figure 2 is a flowchart of the table of contents organization function. Figure 3 is a flowchart of the revision function. The present invention will be described in detail below with reference to the attached drawings. This specification describes only the minimum components necessary for explaining the invention and does not mention components unrelated to the essence of the invention. Furthermore, the mentioned components should not be interpreted in an exclusive sense, but in a non-exclusive sense, which may include other unmentioned components. In this specification, “~part” refers to a logical combination of general-purpose hardware and software that performs the functions thereof. FIG. 1 illustrates a configuration diagram of the book writing software presented by the present invention. The web server unit (100) is composed of a prompt generation unit (101), a database unit (102), a storage unit (103), a communication unit (104), a table of contents configuration API (105), a draft creation API (106), an automatic sentence recommendation API (107), a similar word recommendation API (108), and a revision API (109). The writing unit (200) is composed of an app-based viewer unit (201), a web browser-based viewer unit (202), an input unit (203), a display unit (204), a storage unit (205), and an output unit (206). LLM (300) refers to a machine learning model that generates an appropriate response to user input based on given text data. In particular, LLM (300) can provide higher quality book writing software by using a model that has learned book-related text data. The core APIs (105 to 109) of the book writing software generate text for functions appropriate to the user through LLM (300) and receive and process data through the communication unit (104). Book writing software provides a table of contents organization feature to allow users to easily organize the table of contents. Figure 2 shows a flowchart of the table of contents configuration function performed by the table of contents configuration API (105). The table of contents organization system of the present invention generates and manages a table of contents and sub-tables based on a tree structure. A tree structure is a hierarchical data representation method consisting of a root node and multiple child nodes. This structure is highly suitable for systematically organizing and storing a main table of contents related to the book's subject set by the user, as well as sub-tables dependent on each table of contents. In the tree structure, each node represents a corresponding table of contents or sub-table, and the root node represents the main topic of the book entered by the user. The root node can have multiple child nodes, and each child node represents a sub-table or detailed item. Child nodes can again contain sub-child nodes, allowing for the creation of a multi-level, detailed table of contents structure as needed. For example, the root node (the main topic of the book) could be "Biodiversity," and the first child node (primary table of contents) dependent on it could be "Impacts of Climate Change." Child nodes (secondary table of contents) dependent on "Impacts of Climate Change" could include "Changes in Species" and "Changes in Ecosystems," and in this way, a multi-level ta