KR-20260063060-A - METHOD FOR PROVIDING COMPANY ANALYSIS INFORMATION BASED ON ARTIFICIAL INTELLIGENCE

Abstract

The present disclosure relates to a method and a server for providing AI-based corporate analysis information. An AI-based corporate analysis information provision method performed by a server according to one embodiment of the present disclosure may include: acquiring corporate registration certificate data and corporate-related data from a plurality of sources; identifying registration elements within the corporate registration certificate using an OCR model and a first language model for analyzing the corporate registration certificate data; generating first corporate information data by converting the identified registration elements into a structured form; identifying semantic elements within the corporate-related data using an image recognition model and a second language model for analyzing the corporate-related data; acquiring second corporate information data by converting the semantic elements into a structured form; and generating and providing user-customized corporate analysis information based on user input selecting at least some elements of corporate identification information, the first corporate information data, and the second corporate information data.

Inventors

• 정종현
• 양성권

Assignees

• 주식회사 제제소프트

Dates

Publication Date

20260507

Application Date

20241030

Claims (20)

1. In a method for providing artificial intelligence-based corporate analysis information performed by a server, A step of obtaining corporate registration certificate data and company-related data from multiple sources; A step of identifying registration elements within a corporate registration certificate using an OCR model and a first language model for analyzing the above corporate registration certificate data; A step of generating first corporate information data by converting the identified registration elements into a structured form; A step of identifying semantic elements within the company-related data using an image recognition model and a second language model for analyzing the company-related data; A step of obtaining second corporate information data by converting the above semantic elements into a structured form; and A method comprising the step of generating and providing user-customized corporate analysis information based on user input selecting at least some elements of corporate identification information, the first corporate information data, and the second corporate information data.
2. In paragraph 1, The above-mentioned first language model takes text extracted from the above-mentioned corporate registration certificate data as input and outputs the above-mentioned registration elements, and is a language model fine-tuned using the corporate registration certificate database as training data, and A method in which the second language model receives text extracted from the company-related data as input, outputs semantic elements within the company-related data, and is a language model fine-tuned based on a text database composed of company-related data.
3. In paragraph 1, the above method is, A step of selecting multiple corporations based on user input; and A method further comprising the step of issuing and providing copies of the corporate registration certificates of the plurality of corporations at once within a single process based on the above-mentioned first corporate information data.
4. In paragraph 3, the above method is, A step of generating corporate analysis information corresponding to the plurality of corporations based on the above second corporate information data; and A method further comprising the step of providing a corporate analysis portfolio that manages the aforementioned multiple corporations as a single group.
5. In paragraph 1, the above method is, Steps for identifying updates to the corporate registration certificate of the corporation of interest to the user; and A method further comprising the step of recommending the issuance of an updated corporate registration certificate to the user based on the identification of the above update.
6. In paragraph 5, the above method is, A step of updating the first corporate information data by identifying registration elements within the above-mentioned updated corporate registration certificate; and A method comprising the step of providing information indicating changes before and after the update based on the above-mentioned updated first corporate information data.
7. In paragraph 1, the above method is, A step of collecting a history of user input selecting at least some elements of the first corporate information data and the second corporate information data; and A method further comprising the step of adjusting the order of corporate information elements within the user-customized corporate analysis information based on the history of the user input.
8. In paragraph 7, the above method is, A step of generating a third corporate information data format representing a data structure to be used to generate user-customized corporate analysis information based on a history of user input selecting at least some elements of the first corporate information data and the second corporate information data; and A method further comprising the step of generating and providing user-customized corporate information corresponding to the third corporate information data format based on a user's request to generate new corporate information.
9. In paragraph 8, the above method is, A step of determining weights for corporate information elements corresponding to the third corporate information data format based on the history of the user input; and A method further comprising the step of changing corporate information elements of the third corporate information data format based on the result of comparing each of the above weights with threshold values.
10. In paragraph 9, the above method is, A step of calculating scores for corporate information elements based on the user input history of other users; and A method further comprising the step of determining the threshold values based on scores for the above corporate information elements.
11. In a server that provides AI-based corporate analysis information, Communication interface; Memory that stores one or more instructions; It includes one or more processors that execute one or more instructions stored in the memory, and The above one or more processors, by executing the above one or more instructions, Acquire corporate registration certificate data and company-related data from multiple sources, and Using an OCR model and a first language model for analyzing the above corporate registration certificate data, registration elements within the corporate registration certificate are identified, and The above-mentioned identified registration elements are converted into a structured form to generate first corporate information data, and Using an image recognition model and a second language model for analyzing the above-mentioned company-related data, semantic elements within the above-mentioned company-related data are identified, and The above semantic elements are converted into a structured form to obtain second corporate information data, and A server that generates and provides user-customized corporate analysis information based on user input selecting at least some elements of corporate identification information, the first corporate information data, and the second corporate information data.
12. In Paragraph 11, The above-mentioned first language model takes text extracted from the above-mentioned corporate registration certificate data as input and outputs the above-mentioned registration elements, and is a language model fine-tuned using the corporate registration certificate database as training data, and A method in which the second language model receives text extracted from the company-related data as input, outputs semantic elements within the company-related data, and is a language model fine-tuned based on a text database composed of company-related data.
13. In Paragraph 11, The above one or more processors, by executing the above one or more instructions, Select multiple corporations based on user input, and A server that, based on the above-mentioned first corporate information data, issues and provides copies of the corporate registration certificates of the above-mentioned multiple corporations at once within a single process.
14. In Paragraph 13, The above one or more processors, by executing the above one or more instructions, Based on the above second corporate information data, corporate analysis information corresponding to the above multiple corporations is generated, and A server that provides a corporate analysis portfolio managing the aforementioned multiple corporations as a single group.
15. In Paragraph 11, The above one or more processors, by executing the above one or more instructions, Identify updates to the corporate registration certificates of the corporations of interest to the user, and A server that recommends the issuance of an updated corporate registration certificate to the user based on the identification of the above update.
16. In paragraph 15, The above one or more processors, by executing the above one or more instructions, A step of updating the first corporate information data by identifying registration elements within the above-mentioned updated corporate registration certificate; and A server that provides information indicating changes before and after the update based on the above-mentioned updated first corporate information data.
17. In Paragraph 11, The above one or more processors, by executing the above one or more instructions, Collecting a history of user input selecting at least some elements of the above-mentioned first corporate information data and the above-mentioned second corporate information data, and A server that adjusts the order of corporate information elements within the user-customized corporate analysis information based on the history of the above user input.
18. In Paragraph 17, The above one or more processors, by executing the above one or more instructions, Based on the history of user input selecting at least some elements of the first corporate information data and the second corporate information data, a third corporate information data format representing a data structure to be used to generate user-customized corporate analysis information is generated, and A server that generates and provides user-customized corporate information corresponding to the third corporate information data format based on a user's request to generate new corporate information.
19. In Paragraph 18, The above one or more processors, by executing the above one or more instructions, Based on the history of the above user input, weights for corporate information elements corresponding to the above third corporate information data format are determined, and A server that modifies corporate information elements of the third corporate information data format based on the result of comparing each of the above weights with threshold values.
20. In Paragraph 19, The above one or more processors, by executing the above one or more instructions, Calculate scores for corporate information elements based on the user input history of other users, and A server that determines the threshold values based on scores for the above corporate information elements.

Description

Method for Providing Company Analysis Information Based on Artificial Intelligence The present disclosure relates to a method for providing artificial intelligence-based corporate analysis information, and more specifically, to a method for providing a user with corporate analysis including automatic analysis of a corporate registration certificate and for customizing the items of corporate analysis information for the user. Under current technology, to obtain a Certificate of Incorporation through a company search, users must register with the Korea Internet Registry Office and access the service directly. The Korea Internet Registry Office is a major platform providing Certificates of Incorporation, where users can search for necessary company information and download the certificates. This process is typically performed manually by the user, and the certificates are provided in PDF format. However, there are several issues with the current methods when users seek to analyze corporate information. First, users must personally examine publicly disclosed corporate data or corporate registration certificates to obtain information. This process is time-consuming and labor-intensive, requiring significant manual work to acquire the analysis data users desire. Second, a simple corporate search does not provide access to diverse analysis data. Users require additional analysis tools or information beyond merely downloading registration certificates. Furthermore, registrations downloaded from the online registry are provided as documents in formats such as PDF, meaning they are not machine-readable structured data. This poses limitations to computers in reading the registrations and providing analytical information. While PDF documents are suitable for human reading, they present difficulties for computers to automatically extract and analyze data. Consequently, at the current level of technology, there are limitations in efficiently providing the corporate analysis information desired by users. FIG. 1 is a diagram for schematically explaining the operation of a server according to one embodiment of the present disclosure. FIG. 2 is a flowchart for explaining the operation of a server according to one embodiment of the present disclosure. FIG. 3 is a flowchart illustrating the operation of a server providing a corporate registration certificate according to one embodiment of the present disclosure. FIG. 4 is a flowchart illustrating the operation of a server providing a corporate analysis portfolio according to one embodiment of the present disclosure. FIG. 5 is a flowchart illustrating the operation of a server providing information related to the update of a corporate registration certificate according to one embodiment of the present disclosure. FIG. 6 is a flowchart illustrating the operation of a server adjusting corporate information elements within corporate analysis information according to one embodiment of the present disclosure. FIG. 7 is a flowchart illustrating the operation of a server generating a data format corresponding to user-customized corporate analysis information according to one embodiment of the present disclosure. FIG. 8 is a flowchart illustrating the operation of a server according to one embodiment of the present disclosure applying weights to a data format corresponding to user-customized corporate analysis information. FIG. 9 is a flowchart illustrating the operation of a server according to one embodiment of the present disclosure to determine a threshold value for changing a data format corresponding to user-customized corporate analysis information. FIG. 10 is a block diagram illustrating the configuration of a server according to one embodiment of the present disclosure. In the following, embodiments are described in detail with reference to the attached drawings. However, various modifications may be made to the embodiments, and thus the scope of the patent application is not limited or restricted by these embodiments. It should be understood that all modifications, equivalents, and substitutions to the embodiments are included within the scope of the rights. Specific structural or functional descriptions of the embodiments are disclosed merely for illustrative purposes and may be modified and implemented in various forms. Accordingly, the embodiments are not limited to specific disclosed forms, and the scope of this specification includes modifications, equivalents, or substitutions that fall within the technical concept. Terms such as "first" or "second" may be used to describe various components, but these terms should be interpreted solely for the purpose of distinguishing one component from another. For example, the first component may be named the second component, and similarly, the second component may be named the first component. Furthermore, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless explicitly and specifically defi