Elastic & LG CNS partner to enhance AI search with KeyLook

Elastic has announced a collaboration with LG CNS to enhance the latter's search capabilities with Elasticsearch, resulting in a significant increase in search relevance and faster data retrieval times.

Elastic, known for its Search AI solutions, partnered with LG CNS, a digital transformation company in Korea, to improve their global knowledge management services. This improvement involves integrating Elasticsearch into LG CNS's KeyLook AI, raising the search relevance rate to 95% while halving data retrieval times.

The KeyLook AI search algorithm, developed by LG CNS at its in-house AI Center, is a retrieval augmented generation (RAG) system.

The integration of Elasticsearch is a strategic move to enhance the system's performance and meet the demands of a global market through advanced AI-driven search services.

LG CNS provides consulting, system development, and operational services aimed at driving digital transformation both domestically and globally. The company is known for its expertise in large-scale IT projects and an understanding of generative AI (GenAI) and other advanced technologies.

One of the main objectives for LG CNS in this collaboration was to upgrade the overall search functionality of its KeyLook AI. They aimed to expand the search algorithm's applicability across various fields and languages. A particular challenge was the need to improve search relevancy to comprehend user intent, which required moving from a conventional keyword-based search to a more nuanced context-based vector search approach.

The enhanced search functionality allows users to receive contextually accurate, multilingual results more effectively, according to Vice President Yohan Chin, head of the AI Center at LG CNS. He stated, "With vector search capabilities, Elasticsearch allows for contextually accurate, multi-lingual searches, enhancing the accessibility and usability of corporate knowledge more effectively and efficiently."

The optimisation of memory management and the utilisation of Elasticsearch's scalable architecture were pivotal in halving the data retrieval times. This improvement further supports efficient handling of extensive datasets.

The move to adopt Elasticsearch also benefits LG CNS employees by leveraging GenAI models that encode corporate data into the Elasticsearch system. This allows KeyLook AI to identify related documents and deliver them to a large language model, which then compiles and delivers user-friendly responses to the users.

Matthew Day, General Manager & Vice President, APJ & EMEA at Elastic, commented on the benefits of the collaboration, "Gen-AI powered conversational queries are a powerful tool for next-generation knowledge management, helping to deliver the productivity, efficiency and user experience customers expect. As our collaboration continues with LG CNS, we will continue to explore new avenues of innovation available through LG CNS's KeyLook AI and the Elastic Search AI Platform to benefit our shared customers."