<![CDATA[Customer segmentation is an important aspect of an effective marketing strategy, yet many traditional methods are unable to capture the complexity of diverse customer behaviors. This research aims to apply the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) method for customer segmentation in retail companies, focusing on identifying patterns of purchasing behavior and product preferences. Data was collected through a questionnaire distributed to 500 respondents, then analyzed using the DBSCAN method. The results showed that DBSCAN successfully identified several customer segments with unique characteristics, and provided an average Silhouette Score of 0.67 and Davies-Bouldin Index of 0.45, indicating good cluster quality. The findings imply that a density-based approach can improve a company's understanding of customer dynamics, and enable the development of more targeted and effective marketing strategies. This research makes an important contribution to the marketing literature, while opening up opportunities for further exploration of the use of machine learning methods in customer segmentation.]]>
Copyrights © 2024
