REAL TIME CUSTOMER SEGMENTATION USING HYBRID MODELS FOR SMARTER E COMMERCE

Abstract

E-commerce has experienced rapid growth in India, with revenues surpassing $75 billion in 2023. The Indian e-commerce market is projected to reach $188 billion by 2025, driven by the increasing number of internet users, the rise of digital payments, and the convenience of online shopping. The objective of this project is to develop a robust hybrid unsupervised learning system to identify and segment high- and lowrevenue customers in the e-commerce sector. This segmentation will enable more effective marketing, personalized campaigns, and improved customer retention. Before the advent of machine learning, businesses relied on manual segmentation techniques such as RFM (Recency, Frequency, Monetary) analysis, simple demographic segmentation, or revenue threshold classification based on transactional history. However, traditional customer segmentation methods are limited by static rules and lack the ability to adapt in real time. These approaches often fail to capture dynamic customer behaviors, resulting in inaccurate segmentation, missed revenue opportunities, and diminished customer engagement. As customer data becomes more complex and the need for real-time, precise segmentation increases, manual methods are no longer sufficient. A hybrid unsupervised learning approach addresses this challenge by automatically segmenting customers based on behavioral patterns and transactional data. This enables ecommerce businesses to uncover hidden segments within their customer base, allowing them to target high-value customers with personalized marketing strategies and enhance retention efforts for low-value customers. By leveraging machine learning, particularly hybrid unsupervised models, companies can optimize their revenue strategies and improve overall business performance.