Publicação
Machine Learning Algorithms for E-commerce Personalization: A Systematic Literature Review
| datacite.subject.fos | Ciências Naturais::Ciências da Computação e da Informação | pt_PT |
| dc.contributor.advisor | Rita, Paulo Miguel Rasquinho Ferreira | |
| dc.contributor.advisor | António, Nuno Miguel da Conceição | |
| dc.contributor.author | Sicwebu, Andisiwe | |
| dc.date.accessioned | 2025-11-18T11:51:05Z | |
| dc.date.embargo | 2028-11-03 | |
| dc.date.issued | 2025-11-03 | |
| dc.description | Dissertation presented as the partial requirement for obtaining a Master's degree in Data Driven Marketing, specialization in Digital Marketing and Analytics | pt_PT |
| dc.description.abstract | The increasing reliance on data-driven personalization in e-commerce has positioned machine learning (ML) algorithms at the forefront of enhancing user engagement and driving conversion. However, the variety of algorithmic approaches and the inconsistencies in reported outcomes present a challenge for practitioners and researchers seeking to evaluate their effectiveness. This study conducts a systematic literature review and bibliometric analysis to assess the performance of various ML algorithms used in e-commerce personalization and to map the intellectual structure of the field. Drawing on 32 peer-reviewed articles published between 2014 and 2024, the review compares models such as collaborative filtering, deep neural collaborative filtering (DNCF), and multivariant user interface (UI) personalization based on metrics including click-through rate (CTR), time on site, and conversion rate (CR). The findings indicate that DNCF consistently outperforms traditional models in CTR and recall, while UIdriven strategies show stronger effects on CR. A keyword co-occurrence and citation network analysis further highlight emerging themes, including sentiment analysis, trust modeling, and real-time personalization. This research advances the literature by offering a dual perspective that integrates algorithmic comparison with bibliometric insight, revealing both technical advancements and thematic fragmentation in the field. It provides practical recommendations for e-commerce practitioners while also identifying gaps related to emotional engagement metrics, ethical considerations, and real-time adaptability. The study presents a structured framework for evaluating ML personalization strategies and proposes a future research agenda centered on hybrid systems, fairness, and longitudinal impact assessment. | pt_PT |
| dc.identifier.tid | 204073430 | |
| dc.identifier.uri | http://hdl.handle.net/10362/190934 | |
| dc.language.iso | eng | pt_PT |
| dc.rights.uri | http://creativecommons.org/licenses/by/4.0/ | pt_PT |
| dc.subject | Machine Learning | pt_PT |
| dc.subject | E-commerce Personalization | pt_PT |
| dc.subject | Recommender Systems | pt_PT |
| dc.subject | Systematic Review | pt_PT |
| dc.subject | Bibliometric Analysis | pt_PT |
| dc.subject | SDG 8 - Decent work and economic growth | pt_PT |
| dc.subject | SDG 9 - Industry, innovation and infrastructure | pt_PT |
| dc.subject | SDG 12 - Responsible production and consumption | pt_PT |
| dc.subject | SDG 17 - Partnerships for the goals | pt_PT |
| dc.title | Machine Learning Algorithms for E-commerce Personalization: A Systematic Literature Review | pt_PT |
| dc.type | master thesis | |
| dspace.entity.type | Publication | |
| rcaap.embargofct | I am hoping to publish the research. | pt_PT |
| rcaap.rights | embargoedAccess | pt_PT |
| rcaap.type | masterThesis | pt_PT |
| thesis.degree.name | Mestrado em Marketing Analítico, especialização em Digital Marketing and Analytics | pt_PT |