Jun Ji-hye, a reporter at the finance desk of The Korea Times, focuses primarily on economic policy and government agencies, mainly covering the Ministry of Finance and Economy, the Ministry of Budget and Planning, the National Tax Service and the Korea Customs Service. She previously covered financial authorities, including the Financial Services Commission and the Financial Supervisory Service, and earlier worked on the political, city and business desks, reporting on a wide range of issues.
Kbank showcases AI-powered personalization model in academic journal
Kbank's headquarters in Seoul / Courtesy of Kbank
By Jun Ji-hye
A research paper authored by Kbank employees detailing the bank's personalized recommendation system powered by artificial intelligence (AI) has been published in a leading academic journal in Korea, the bank said Monday.
The publication reflects academic recognition of the company's research capabilities in AI technology, it added.
The paper appeared in the Journal of the Korean Data Analysis Society (JKDAS), published by the Korean Data Analysis Society (KDAS). The journal is one of Korea's major academic publications and listed in the Korea Citation Index (KCI).
Kwon Hyuk-min, Lee Sang-hyun and Cho Yong-geol of the internet-only bank's data intelligence team co-authored the paper, titled "Strategic Design of an AI-Based Recommendation System and Its Impact on User Experience: An MLOps-Driven Experimental Study in a Financial App."
MLOps stands for machine learning operations.
The paper presents an empirical study on the impact of the bank's AI-powered personalized recommendation system on customer behavior, user experience and overall business performance.
The research is considered a meaningful contribution, as it explores AI-driven personalization strategies in the financial sector, where such approaches remain relatively uncommon compared to industries like e-commerce and streaming services.
In a bid to develop a model tailored specifically to the needs of the financial sector, Kbank has conducted focus group interviews with internal experts from various divisions, including lending and deposits.
These discussions allowed the bank to gain deeper insights into customer segments and behavioral trends, which were integrated into the AI model from the initial development stage.
By incorporating these insights, the system delivers recommendations that reflect the specific characteristics of financial consumers, going beyond purely technology-driven suggestions. This approach has led to improvements in both predictive accuracy and system stability.
"We will continue to advance our in-house AI system into a comprehensive AI agent framework to provide customers with more refined and personalized financial services," a Kbank official said.