Hybrid Clustering with Deep Learning in E-commerce for Customer Segmentation: A Data-Driven Approach for Business Strategy Optimization

Abstract

Customer segmentation is a strategic approach to understanding customer
needs and preferences, especially in the dynamic e-commerce industry.
Traditional clustering methods, such as k-means, are often used for this task,
but have limitations in handling complex and high-dimensional data. In this
research, we use a hybrid clustering approach that integrates deep learning for
feature extraction with traditional clustering algorithms for customer
segmentation. Uses Mall Customers Dataset from Kaggle, which includes
customer demographic and shopping behavior data. Experimental results
show that this approach is able to produce more accurate and meaningful
segmentation. The visualization of the results shows significant patterns that
can be used to develop more personalized and effective marketing strategies.