Study of Awareness Patterns of Credit Card Users towards Ads with K-Means Clustering Algorithm

Abstract

In the era of digital transformation, credit cards have become an essential component of modern financial life, where users’ understanding of card features significantly influences their financial decisions. Despite the wide use of advertising in the financial sector, limited studies have explored how credit card users in emerging markets respond to such campaigns. Addressing this gap, this study analyzes advertisement awareness patterns among credit card users in Indonesia using the K-Means Clustering algorithm on a dataset collected from August 2023 to March 2024. The study aims to examine levels of advertisement awareness, segment users based on their responses, and assess the implications of these segments for marketing strategies. The methodology follows the Knowledge Discovery in Database (KDD) process: data selection, preprocessing, transformation, clustering with K-Means, and evaluation using the Silhouette Score. Results reveal three distinct user clusters: (1) highly aware users in large cities with high exposure; (2) moderately aware users from mid-tier cities; and (3) low-awareness users despite high exposure, often from older age groups and lower SES backgrounds. The clustering yielded Silhouette Scores above 0.60, validating segmentation quality. The novelty lies in applying machine learning to segment awareness levels using a multi-city real-world dataset. The findings offer practical value for credit card providers to enhance targeted campaigns, improve user engagement, and allocate marketing resources more effectively across demographic segments.