Challenge
Public construction data—such as text, images, and other materials—is scattered across various institutions, limiting accessibility and practical use. As a non-profit organization, there is a need for a service that leverages the existing platform to identify and prioritize key information, providing tailored insights that help companies make informed decisions when entering overseas construction markets.
Approach
An intelligent chatbot service powered by Retrieval-Augmented Generation (RAG) was developed to deliver practical, high-quality answers based on an internal database. In addition to existing data, the system rapidly incorporates large volumes of publicly available search content, enabling accurate and responsive service delivery within a short period.
Value Delivered
An intelligent search service was implemented on a web-based platform to improve access to public data. The service enhanced usability by providing diverse answers derived from both real-time search content and the existing database. To ensure long-term scalability, it was designed to support continuous training from newly acquired data—enabling the development of a reliable and scalable AI service.
