Online Retail Big Data Analysis Using H2O and Machine Learning

The exponential growth of online retail in recent years has demanded businesses to analyze online retail data to comprehend customer behavior and enhance productivity. In this paper, a new framework is proposed that can be used for customer segmentation and customer prediction. Specifically, clustering techniques are employed, an unsupervised learning technique, for customer segmentation based on Recency, Frequency, and Monetary (RFM) parameters. Moreover, a comprehensive analysis of an online retail dataset is conducted, leveraging techniques from big data mining and machine learning. Subsequently, appropriate labels are assigned to each customer segment and develop classification models such as decision trees, random forests, and k-nearest neighbors. These models enable the prediction of future customer behavior, facilitating targeted marketing strategies and personalized customer experiences. By harnessing the power of big data analytics, this paper contributes to the ongoing discourse on enhancing customer understanding and driving business growth in the dynamic landscape of online retail.