Implementasi Data Mining dalam Identifikasi Pola Perilaku Konsumen pada Platform E-Commerce

Abstract

The rapid growth of digital technology has significantly accelerated the expansion of the e-commerce sector, generating large and continuously increasing volumes of customer transaction data. This data contains valuable insights that can be utilized to better understand consumer behavior. Data mining is a widely used approach to extract such insights by identifying patterns, trends, and customer characteristics based on transaction activities. This study aims to implement data mining techniques to identify consumer behavior patterns on e-commerce platforms in order to support more effective and targeted marketing strategies. The research method involves collecting customer transaction data, including purchase frequency, transaction value, product categories, transaction time, and customer interaction levels. The data is then processed through cleaning, transformation, and normalization to ensure quality. K-Means clustering is applied to group customers based on similar behavioral characteristics, while the Apriori algorithm is used to analyze associations between products frequently purchased together. The results identify distinct customer segments, including loyal customers, moderate buyers with specific preferences, and sporadic low-value buyers. Association analysis also reveals product bundling patterns, supporting cross-selling and recommendation strategies, and improving overall marketing effectiveness.