LG CNS Consortium Selected as Preferred Bidder for Gyeonggi Education Office Digital Platform
Largest AI Project Commissioned by a Public Institution This Year
Development of AI Assistant for Staff and Educational Data Analysis Platform
LG CNS announced on June 23 that it has been selected as the preferred bidder for the Gyeonggi Provincial Office of Education's "AI-Data-Centric Gyeonggi Education Digital Platform" project.
This project is valued at a total of 38 billion KRW, making it the largest artificial intelligence (AI) project commissioned by a public institution this year.
LG CNS has formed a consortium with NHN INJEINC, Softitech, and SGA to participate in the project. The LG CNS consortium outperformed the Samsung SDS-KT consortium in the technical evaluation.
This project aims to build an intelligent digital platform based on generative AI to comprehensively support the educational operations of the Gyeonggi Provincial Office of Education for students, parents, and staff. LG CNS will be responsible for the AI and data-based digital platform, generative AI services, automated data collection system, AI data collection and management system, and cloud infrastructure setup.
Through the establishment of this platform, educational sites will be provided with an "AI Administrative Assistant" that supports staff with tasks such as drafting documents like home correspondence and providing report summarization and translation services. An "AI Communication Assistant" will deliver various school life information to students and parents. An AI copilot service will also be offered to integrate the management of email, messenger, meetings, and schedules.
An educational data analysis platform will also be developed for personalized student guidance and educational policy formulation. To achieve this, a system will be established to automatically collect data generated from various school websites and the Office of Education's business systems.
LG CNS will apply Retrieval-Augmented Generation (RAG) technology to meet the needs of educational environments, where revised laws and the latest guidelines must be reflected in daily operations. RAG is a technology that combines external knowledge search capabilities not included in the AI model's training data to improve the response accuracy of generative AI.
LG CNS plans to utilize the "Exaone" large language model (LLM) developed by LG AI Research for this project. By applying LG CNS's fine-tuning technology to Exaone and training it with information from the Gyeonggi Provincial Office of Education, the company aims to develop an AI model optimized for educational settings. Fine-tuning is a technique that specializes an LLM for specific purposes by training it with separate datasets.
LG CNS plans to complete the construction of the "AI-Data-Centric Gyeonggi Education Digital Platform" and introduce AI services optimized for educational environments by June next year.
© The Asia Business Daily(www.asiae.co.kr). All rights reserved.
