The Journal of Machine Learning Research
Spectral clustering for multi-type relational data
ICML '06 Proceedings of the 23rd international conference on Machine learning
Segmenting Customers from Population to Individuals: Does 1-to-1 Keep Your Customers Forever?
IEEE Transactions on Knowledge and Data Engineering
Journal of Management Information Systems - Special section: Data mining
Expert Systems with Applications: An International Journal
Classifying the segmentation of customer value via RFM model and RS theory
Expert Systems with Applications: An International Journal
Electronic Commerce Research and Applications
Electronic Commerce Research and Applications
Review: Soft computing applications in customer segmentation: State-of-art review and critique
Expert Systems with Applications: An International Journal
Expert Systems with Applications: An International Journal
Hi-index | 0.00 |
The segmentation of online consumers into multiple categories can contribute to a better understanding and characterization of purchasing behavior in the electronic commerce market. Online shopping databases consist of multiple kinds of data on customer purchasing activity and demographic characteristics, as well as consumption attributes such as Internet usage and satisfaction with services. Information about customers uncovered by segmentation enables company administrators to establish good customer relations and refine their marketing strategies to match customer expectations. To achieve optimal segmentation, we developed a soft clustering method that uses a latent mixed-class membership clustering approach to classify online customers based on their purchasing data across categories. A technique derived from the latent Dirichlet allocation model is used to create the customer segments. Variational approximation is leveraged to generate estimates from the segmentation in a computationally-efficient manner. The proposed soft clustering method yields more promising results than hard clustering and greater within-segment clustering quality than the finite mixture model.