Loading...

Clustering and Analyzing Online Business Customer Behavior using Ensemble Learning Methods

Mokaffeli Shiramin, Ali | 2025

0 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 58575 (01)
  4. University: Sharif University of Technology
  5. Department: Industrial Engineering
  6. Advisor(s): Hassan Nayebi, Erfan
  7. Abstract:
  8. In the age of information and technology, online stores have significantly expanded as one of the most prominent manifestations of e-commerce. With increasing competition among businesses, leveraging modern data mining techniques to identify potential customers, predict customer churn, and enable more precise targeting in direct marketing has become a necessity. This study integrates data mining methods with marketing concepts to analyze the behavior of online store customers using the RFM model (Recency, Frequency, and Monetary value of purchases) and employs the K-Means clustering algorithm to segment customers. Furthermore, to more accurately predict customer behavior, two modeling approaches have been developed: one to identify and predict loyal customers and another to detect customers at risk of churn. In this regard, an ensemble learning model based on the StackingClassifier algorithm has also been developed to improve the accuracy of these predictions, especially for loyal customers. Experimental results showed that the proposed ensemble learning model outperformed individual models and previous studies, increasing prediction accuracy for identifying loyal customers by up to 11.8%. This research also presents a model for identifying customers on the verge of churning, which, alongside the loyal customer prediction model, provides a powerful tool for marketing managers. This significant improvement in predicting loyal customers confirms the high efficiency of the proposed hybrid model in customer behavior analysis. In addition to offering more accurate models, this study introduces a new perspective on customer data analysis, playing a crucial role in intelligent marketing decision-making, developing customer retention strategies, enhancing customer engagement, and improving recommender systems. The findings of this research can serve as a foundation for optimizing loyalty programs, designing preventive actions to avoid customer churn, and personalizing services in digital businesses.
  9. Keywords:
  10. Customer Clustering ; Data Mining ; K-means Clustering ; Customer Profiling ; Recency, Frequency and Monetary Value of Purchases (RFM)Model ; Online Store

 Digital Object List

 Bookmark

No TOC