Effectiveness of Uplift Modeling When Multiple Treatments are Tested in Fashion E-Commerce Campaigns

Konji, Prajwal Keshav (2024) Effectiveness of Uplift Modeling When Multiple Treatments are Tested in Fashion E-Commerce Campaigns. Masters thesis, Dublin, National College of Ireland.

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

Abstract

The fashion e-commerce sector relied on generalized marketing methods like sending promotional messages to every customer, leading to customer fatigue and ineffective campaigns. This study aims to apply and evaluate multiple uplift modeling techniques to target potential customers in the e-commerce domain. The focus is on identifying ’Persuadable’ customers, those who are likely to respond positively to marketing messages sent through SMS, Email, or Push Notifications. The study is done on a Russian medium-sized online fashion retailer and represents customer’s interaction with campaigns from 2021-2023. The message clicked by the customer is considered the target action in this study. Four uplift models are implemented, the Two-Model Approach with CatBoost and LightGBM classifiers, Class Transformation, and T-Learner with LightGBM. The metrics used for the evaluation are Uplift Score at 30%, AUUC, and AUQC. Class Transformation achieved the best Uplift Score at 30% with 7%, indicating moderate success in targeting potential customers. Generally, the overall performance across the models was mixed, with low AUUC and AUQC scores. Signifying challenges in generalizing the models across the data. The research suggests uplift modeling be applied to enhance targeted marketing in fashion e-commerce. However, the study showed several key limitations such as computational complexity, and difficulties with capturing complex customer behaviors.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Mulwa, Catherine 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 > Electronic Commerce H Social Sciences > HD Industries. Land use. Labor > Specific Industries > Fashion Industry 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 09:39
Last Modified:	20 Aug 2025 09:39
URI:	https://norma.ncirl.ie/id/eprint/8582

Actions (login required)

View Item