A Smart Model to enhance Customer retention in E-commerce | ||||
| المجلة العلمية للبحوث والدراسات التجارية | ||||
| Volume 38, Issue 1, March 2024, Page 1501-1538 PDF (573.99 K) | ||||
| Document Type: المقالة الأصلية | ||||
| DOI: 10.21608/sjrbs.2024.270306.1629 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
Norihan N. Abdelgawad  1; Aya M. Mostafa 2
| ||||
| 1Teaching Assistant in October High Institute for Engineering & Technology. | ||||
| 2Assistant Professor of Information Systems, Information Systems Department, Faculty of Commerce and Business Administration, Helwan University, Cairo, Egypt. | ||||
| Abstract | ||||
| Customer relationship management is a tool that helps improve customer retention, which is the foundation of any business and can be targeted on churn prediction, as early and effective detection of customer churn is necessary for supporting the organization as a whole. Due to the exponential growth of data volume in e-commerce, it is essential to develop techniques to identify customer churn. Machine learning and CRM will effectively perform this task and result in changes in business environments. Additionally, customer retention strategies can focus on rush time prediction within an e-commerce platform. This approach emphasizing that employee distribution and minimizing technical issues will not only enhance operational efficiency but also lead to increased customer satisfaction. This paper aims to achieve customer retention by incorporating both customer churn prediction and the prediction of rush times for purchases. The first part of the experiment employed K-means customer segmentation based on RFM model with various machine learning algorithms, including Decision Tree, Random Forest, and XG-Boost, to predict customer churn. we addressed imbalanced data by combining the Synthetic Minority Over-sampling Technique with Tomek links (SMOTE-TOMEK). The results indicated that XG-Boost model outperformed other algorithms, achieving an accuracy of 87.37%, precision of 99.99%, recall of 74.68%, F1 score of 85.5%, and an AUC of 88%. The second part of the experiment employed z-scores to analyze the timing of customer purchases. The results indicated that the peak purchasing period, defined as the time between 4:57 AM and 6:04 PM, included approximately 68% of all purchases. | ||||
| Keywords | ||||
| Customer retention; E-Commerce; Ensemble algorithms | ||||
|
Statistics Article View: 418 PDF Download: 314 |
||||
