Enhancing Purchase Predictions with Machine Learning: Customer Propensity Modelling through Predictive Analytics

Gudibanda Prasnnakumar, Vinith Kumar (2023) Enhancing Purchase Predictions with Machine Learning: Customer Propensity Modelling through Predictive Analytics. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (929kB) | Preview
Preview	PDF (Configuration Manual) Download (3MB) | Preview

Abstract

In this research, the aim was to predict online customer purchasing behavior accurately. This issue is essential to businesses as they seek to optimize marketing efforts and improve customer engagement by determining the likelihood of purchase transactions. However, traditional analytics has limitations when dealing with imbalanced datasets and complex patterns in customer behavior, which necessitated this research. One of the limitations identified was that research indicates that the typical conversion rate in e-commerce is only about 2-3%. To address this issue, a dataset was sourced from the UCI repository, and a novel approach of using multiple algorithms such as Logistic Regression, Decision Tree classifier, GBM, and RNN was adopted. The model's ability to identify confirmed transactions (marked by the 'Revenue' class) was significantly improved by incorporating advanced data preprocessing techniques such as SMOTE, OLS, and RFECV. The technical approach included a detailed analysis of the dataset, pre-processing methods to improve data quality, and the use of GBM to model purchasing behavior. To ensure efficient performance across different data scenarios, GBM parameters were optimized using a rigorous cross-validation process, which is the novelty of this research. The results showed that when combined with our pre-processing strategy, the GBM outperforms standard prediction models. The accuracy of identifying 'Revenue' class transactions has significantly improved, providing businesses with actionable insights into customer purchasing patterns. The research identifies key factors that influence customer decisions, allowing for the development of more targeted and effective marketing strategies.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Rustam, Furqan UNSPECIFIED
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 H Social Sciences > HF Commerce > Marketing
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	08 May 2025 14:14
Last Modified:	08 May 2025 14:14
URI:	https://norma.ncirl.ie/id/eprint/7519

Actions (login required)

View Item