"Clustering and Predictive Analysis of Fourth-Generation K-Pop Album Sales Success Using Data Mining Algorithms"

Abstract

Abstract

The rapid development of South Korea's music industry, particularly in the fourth generation of K-Pop, has shown a significant increase in both physical and digital album sales. This study aims to explore the sales patterns of fourth-generation K-Pop artists based on cross-country sales data, focusing on data modeling and pattern analysis using data mining techniques. The dataset used consists of 97 album sales entries from several fourth-generation K-Pop artists, including information such as artist, album title, release date, country, sales volume, and peak chart position. Through data cleaning, feature processing, and the application of algorithms such as K-Means Clustering and Random Forest Classifier, this research provides insights into the dominant factors influencing sales performance and the clustering patterns of artists based on market and chart achievements. The results offer strategic information about K-Pop market behavior and can serve as a basis for the music industry to make more effective business decisions.