KR-20260062385-A - SYSTEM FOR POLICY INTELLIGENCE SERVICE PROVISION BASED ON DIGITAL TWIN AND METHOD USING THEREOF

Abstract

The present invention relates to a digital twin-based high-reliability policy intelligence service provision system and a method thereof. A digital twin-based policy intelligence service providing system according to the present invention includes a reliability management module that identifies a response type and a reliability requirement level for a user's request, a response generation module that generates a response corresponding to the response type and the reliability requirement level, and a response service providing module that provides the response to the user.

Inventors

• 이연희
• 강현중
• 김영민
• 김태환
• 김현재

Assignees

• 한국전자통신연구원

Dates

Publication Date

20260507

Application Date

20241029

Claims (19)

1. A reliability management module that identifies response types and reliability requirement levels for user requests; A response generation module that generates a response corresponding to the above response type and reliability requirement level; and Response service provision module that provides the above response to the user A digital twin-based policy intelligence service provision system including
2. In paragraph 1, The above reliability management module includes a request determination unit that determines the response type and required reliability level for the user's request, a response reliability determination unit that determines the reliability of the generated response, and a model reliability level determination unit that determines the reliability satisfaction level for the model used to generate the response. Digital Twin-based Policy Intelligence Service Provision System.
3. In paragraph 1, Further including an instance creation module that provides time-specific information required for the generation of the above response. Digital Twin-based Policy Intelligence Service Provision System.
4. In paragraph 3, Further including an active data collection module that collects data to provide the above-mentioned information at each point in time. Digital Twin-based Policy Intelligence Service Provision System.
5. In paragraph 1, Including an additional model optimization module to optimize model parameters Digital Twin-based Policy Intelligence Service Provision System.
6. In a method for providing digital twin-based policy intelligence services performed by a digital twin-based policy intelligence service provision system, (a) A step of receiving a request from a client for a forecast or analysis service regarding a specific point in time; (b) a step of interpreting a service request received from the client and determining the response type and response reliability; (c) a step of generating instances according to the above response type and response reliability level; (d) generating a response from the instance according to the level of the response confidence level; and (e) providing a response service to the client according to the level of the response reliability. A method for providing digital twin-based policy intelligence services including
7. In paragraph 6, The above step (b) is to identify the time point to be analyzed and to determine whether it is necessary to provide the basis for the predicted value and to suggest alternatives. Method for providing digital twin-based policy intelligence services.
8. In paragraph 6, Step (b) above determines the level of response reliability by identifying the degree of demand for the realistic explanatory power of the response and the interpretability of the model. Method for providing digital twin-based policy intelligence services.
9. In paragraph 8, Step (b) above determines the level of response reliability based on an explicit response reliability requirement, or determines the level of response reliability using requester characteristic information or outlook clock information. Method for providing digital twin-based policy intelligence services.
10. In Paragraph 9, Step (b) above grants bonus points for the reality explanatory power or model interpretability when the requester characteristic information is used. Method for providing digital twin-based policy intelligence services.
11. In paragraph 6, The above step (c) involves training a report generation model using digital twin instances recreated at each time point and collected past reports at each time point, and generating digital twin instances at the target time point. Method for providing digital twin-based policy intelligence services.
12. In Paragraph 11, Step (d) above involves generating the response in the form of a report at a specific point in time using the report generation model trained with the digital twin instance at the target point in time as input. Method for providing digital twin-based policy intelligence services.
13. In Paragraph 11, Step (c) above involves creating a digital twin monitoring instance of a past point in time if the response type is an analysis report on a past point in time, running a simulation to create a digital instance evolved one quarter if the response type is a secondary forecast report, and running a simulation up to a medium-to-long-term target point in time to create a digital twin instance for each evolutionary quarter if the response type is a medium-to-long-term forecast report. Method for providing digital twin-based policy intelligence services.
14. An input interface device that receives a forecast or analysis service request from a client; Memory where a digital twin-based policy intelligence service provider program is stored; and It includes a processor that executes the above program, The processor interprets a request received from the client, determines the response type and response reliability, and generates a response. Digital twin-based policy intelligence service provider.
15. In Paragraph 14, The processor determines the level of response reliability by identifying the required degree of the response's explanatory power and the model's interpretability. Digital twin-based policy intelligence service provider.
16. In paragraph 15, The above processor assigns bonus points to the reality explanatory power and model interpretability using requester characteristic information. Digital twin-based policy intelligence service provider.
17. In Paragraph 14, The above processor generates time-series information necessary to generate the above response. Digital twin-based policy intelligence service provider.
18. In Paragraph 17, The above processor trains a report generation model using digital twin instances recreated at each point in time and collected past reports at each point in time, and generates a digital twin instance at a target point in time. Digital twin-based policy intelligence service provider.
19. In Paragraph 18, The processor generates the response in the form of a report at a specific point in time using the report generation model trained with the digital twin instance at the target point in time as input. Digital twin-based policy intelligence service provider.

Description

System for Policy Intelligence Service Provision Based on Digital Twin and Method Using Thereof The present invention relates to a digital twin-based high-reliability policy intelligence service provision system and a method thereof. Since decision-making in the government and public policy sectors has significant national repercussions, there is a need for technologies that support decision-making to ensure sound decisions. There is a demand for AI-based decision support technologies that possess accuracy and transparency, such as making correct judgments on real-world economic phenomena, accurate predictions of the future, and presentation of policy alternatives; however, no concrete measures capable of meeting these demands have been presented. FIG. 1 illustrates a digital twin-based policy intelligence service provision system according to an embodiment of the present invention. FIG. 2 illustrates a method for providing a digital twin-based policy intelligence service according to an embodiment of the present invention. FIG. 3 illustrates the process of determining the response type and response reliability level of a request determination unit according to an embodiment of the present invention. FIG. 4 illustrates a response generation process according to a response type and response reliability level according to an embodiment of the present invention. FIG. 5 illustrates a reinforcement learning-based simulation model optimization process for deriving a basis scenario for an inference result using a prediction model according to an embodiment of the present invention as a reference model. FIG. 6 illustrates the process of deriving a basis scenario based on reinforcement learning and simulation models to generate an explanation for a specific point in time according to an embodiment of the present invention. FIG. 7 illustrates a process for determining the reliability level of a model according to an embodiment of the present invention. FIG. 8 illustrates a procedure for generating a response in the form of a digital twin-based report according to an embodiment of the present invention. FIG. 9 illustrates a digital twin-based response generation process for a request to generate a medium-to-long-term economic outlook report according to an embodiment of the present invention. FIG. 10 illustrates an active digital twin improvement process according to an embodiment of the present invention. FIG. 11 illustrates a digital twin-based active service device according to an embodiment of the present invention. FIG. 12 is a block diagram showing a computer system for implementing a method according to an embodiment of the present invention. The aforementioned objectives of the present invention, as well as other objectives, advantages, and features, and the methods for achieving them, will become clear from the embodiments described in detail below together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below but can be implemented in various different forms, and the following embodiments are provided merely to easily inform those skilled in the art of the purpose, structure, and effects of the invention, and the scope of the rights of the present invention is defined by the description in the claims. Meanwhile, the terms used in this specification are for describing the embodiments and are not intended to limit the invention. In this specification, the singular form includes the plural form unless specifically stated otherwise in the text. As used in this specification, "comprises" and/or "comprising" do not exclude the presence or addition of one or more other components, steps, actions, and/or elements to the mentioned components, steps, actions, and/or elements. In the following, the background of the proposed invention is explained to aid the understanding of those skilled in the art, followed by an explanation of embodiments of the invention. Since decision-making in the government and public policy sectors carries significant national repercussions, there is a need for technologies that support decision-making to ensure sound decisions. While there is a demand for AI-based decision support technologies that possess accuracy and transparency—such as making sound judgments on real-world economic phenomena, accurately predicting the future, and presenting policy alternatives—concrete measures capable of meeting these demands have not yet been presented. According to conventional technology, traditional statistical models, time series forecasting models, and low-resolution simulation-based models are utilized, often resulting in poor explanatory power regarding reality, such as producing results that are detached from reality. While training a model with a large number of parameters from various datasets, such as deep learning, can learn relatively complex data patterns and achieve high prediction performance, there are problems such as reduced tr