Personalized Student Information Support via Web Chatbot

Overview

School environments face challenges where faculty members struggle to focus on core educational activities due to a high volume of simple, repetitive inquiries* from students and parents.
*Academic/event schedules, attendance/field trip application procedures, facility usage, school newsletters, etc.

Currently, education-related information is fragmented across multiple channels, forcing users to navigate various services to find information. This has increased the administrative burden of responding to repetitive complaints. To address this, the project aimed to build a customized guidance system at the school level using Hyper-scale AI technology.

Project description

The project collected data from educational service materials and website announcements for students and parents. Requirements were analyzed through user interviews to derive improvements for service enhancement. Additionally, this project explored cloud-based system configurations and AI model optimization to develop a strategic plan for a "Educational Community Portal."

By collecting rule-based chatbot Q&A data (including scenarios) and institutional website data from the Gyeonggi Provincial Office of Education, the project implemented data preprocessing, embedding, and Vector DB construction. Finally, this service established a Generative AI chatbot that responds to various simple inquiries by integrating Hyper-CLOVA X APIs with optimized prompting.

Key technological features

• Hyper-scale AI Integration: Implementing customized Q&A services by adopting Hyper-scale AI chatbot technology to ensure personalized and human-like interactions
• Advanced RAG Architecture: Building a Retrieval-Augmented Generation (RAG) framework based on Hyper-CLOVA X (LLM), structured Q&A, and rule-based scenarios to maximize response accuracy

Expected outcomes

Through this Generative AI chatbot pilot project, the service provides an integrated and optimized Q&A system for a wide range of education-related inquiries. By offering 24/7 consultation and customized information in a single location, the project enables students and parents to receive immediate responses, thereby significantly enhancing user satisfaction. Furthermore, this service is expected to create an environment where faculty members can focus more effectively on core educational activities by automating simple, repetitive inquiries and reducing their overall administrative workload.

Launch year: 2024

Attribution

Case study courtesy of the National Information Society Agency, Republic of Korea, Apolitical's Content Partner. Sourced from Use Cases of Public AI Service, Vol.2, https://eng.nia.or.kr/site/nia_eng/ex/bbs/View.do?cbIdx=31975&bcIdx=29182&parentSeq=29193