Advanced Predictive Modelling of E-Commerce Customer Behaviour: Integrating Machine Learning and Deep Learning Techniques

Meka, Venkata Naveen (2024) Advanced Predictive Modelling of E-Commerce Customer Behaviour: Integrating Machine Learning and Deep Learning Techniques. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (1MB) | Preview

Abstract

As the sector of e-commerce have been evolving drastically in recent years, understanding and predicting customer behavior has become essential for business owners. This research aims to overcome the issues involved in e-commerce through predicting and analyzing purchases, cart abandonment, understanding the seasonality impact on conversion rates, carrying out an in-depth clickstream analysis, customer lifetime value (CLV) estimation and lastly, predicting future monthly sales. These concerns have a huge impact on income generation and customer retention. However, they are challenging because of the sophisticated and dynamic nature of online buying behaviours. This research is motivated by the critical requirement to improve predictive analytics in e-commerce. Doing so can lead to more personalised and successful marketing strategies, resulting in higher conversion rates and increased consumer success. The dataset used for this study is "Online Shopper's Intention", provided on the UCI Machine Learning Repository. In this research, supervised learning methods such as Random Forest, XGBoost, and Logistic Regression were used for purchase prediction, and deep learning models including an ensemble of Long Short-Term Memory (LSTM-RF) model, and Bi-LSTM were developed for predicting and analysing cart abandonment. These models were improved using hyperparameter tuning, and then it proceeded to test for performance based on the metrics including accuracy, precision, recall, F1 score, and ROC-AUC. Checking if seasonality affects the conversion rates involved analyzing weekends, special days like bank holidays, months and other factors effect on revenue. An in-depth clickstream data analysis was performed to see if the time spent on a particular page has an effect on the conversion rate. Customer Lifetime Value (CLV) analysis was performed to understand about customer retention and Time Series Analysis was performed to predict future monthly sales. The results of these analysis provide business owners good insights to better understand the intricate nature of customer behaviour, to carry out personalised marketing strategies that will increase customer satisfaction and the overall revenue generation.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Basilio, Jorge UNSPECIFIED
Uncontrolled Keywords:	e-Commerce Optimization; Purchase Intention Prediction; Cart Abandonment; Seasonality Impact; Clickstream Data Analysis; User Behaviour Patterns
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science H Social Sciences > HF Commerce > Marketing > Consumer Behaviour H Social Sciences > HF Commerce > Electronic Commerce Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	20 Aug 2025 10:26
Last Modified:	20 Aug 2025 10:26
URI:	https://norma.ncirl.ie/id/eprint/8587

Actions (login required)

View Item