Segmentasi Pengguna Spotify Berdasarkan Preferensi Musik Dengan Algoritma K-Means Clustering

Abstract

Music streaming platforms like Spotify have become integral to the daily lives of millions globally, offering personalized listening experiences. However, managing a vast music catalog to present relevant content to each user remains a challenge. This study explores the application of the K-means clustering algorithm to segment Spotify users based on their music preferences. The goal is to group users into clusters with similar tastes to enhance targeted marketing and user engagement. We utilized a secondary dataset of trending Spotify songs and their attributes from 2023. Through data preprocessing, feature selection, and normalization, we prepared the data for clustering. The optimal number of clusters was determined using the Elbow Method, resulting in six distinct clusters. Each cluster represents unique music preferences, analyzed through metrics such as danceability, energy, and popularity. The findings demonstrate that K-means clustering effectively identifies user segments, providing insights for improving personalized recommendations and marketing strategies. This research underscores the potential of machine learning in optimizing user experiences on music streaming platforms.

Keywords: Spotify, Elbow Method, K-Means Clustering, Music, User Segmentation