Characterising Online Purchasing Behaviour

EasyChair Preprint 10460

Abstract

This paper suggests a system that classifies customers based on their purchase behaviour and examines their patterns using supervised and unsupervised learning techniques. The technology analyses consumer behaviour at the session and user journey levels, anticipating consumer behaviour with excellent accuracy and recall rates. Five main client clusters with different behaviour patterns were identified by the study: "new customers," "active shoppers," "returning decisive shoppers," "comparison shoppers," and "window shoppers." These patterns can offer insightful information for developing personalised marketing tactics, enhancing the entire consumer experience, boosting client retention, and promoting business expansion. The study emphasises the value of client segmentation and its possible commercial advantages.

Keyphrases: AutoML, Clustering, Random Forest, machine learning, supervised learning, unsupervised learning

@booklet{EasyChair:10460,
  author    = {Ganesh Gurram and Nikhil Reddy Kodumuru and Mukesh Tripathi},
  title     = {Characterising Online Purchasing Behaviour},
  howpublished = {EasyChair Preprint 10460},
  year      = {EasyChair, 2023}}