RFMVDA: An Enhanced Deep Learning Approach for Customer Behavior Classification in E-Commerce Environments
Customer Relationship Management (CRM) systems, widely used in enterprises, have evolved into Software-as-a-Service (SaaS) platforms. With the advent of Customer Data Platforms (CDP), these systems continuously store customer behavior data for purposes such as creating single customer profiles, anal...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10839360/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Customer Relationship Management (CRM) systems, widely used in enterprises, have evolved into Software-as-a-Service (SaaS) platforms. With the advent of Customer Data Platforms (CDP), these systems continuously store customer behavior data for purposes such as creating single customer profiles, analyzing, tracking, and managing customer interactions from various perspectives. With the global expansion of the e-commerce market, research on customer analysis and classification optimized for the e-commerce environment has been actively conducted. The RFM (Recency, Frequency, Monetary) model is a straightforward method for classifying customers and is applied across various industries. However, in the e-commerce environment, where customers can access services at any time, there are limitations in collecting, storing, and reflecting customer behavior data for classification. To resolve these limitations, this paper proposes the RFMVDA (Recency, Frequency, Monetary, Visits, Durations, Actions) model. This model is designed to capture customer data, sessions, and behavior units suitable for the e-commerce environment. By utilizing the RFMVDA model for customer behavior-based segmentation and classification, we constructed a Deep Neural Network (DNN) to predict customer behavior-based classifications. As a result, the proposed model demonstrated a segmentation prediction accuracy of 92.98% for customers in the e-commerce environment. |
|---|---|
| ISSN: | 2169-3536 |