Optimizing Machine Learning Models for Real-Time Detection of Fake Product Reviews on E-commerce

Chollety, Manoj Kumar (2025) Optimizing Machine Learning Models for Real-Time Detection of Fake Product Reviews on E-commerce. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (870kB) | Preview

Abstract

Fake reviews are a significant challenge on e-commerce platforms, due to the lack of trust and fair play with consumers. This project is aiming at creating a machine learning based automatic fake review detection system. A cleaning of the incoming datasets; The OR, and CG were subjected to clean outliers, data wrangle followed by an exploratory data analysis. I used the TF-IDF (Term Frequency – Inverse Document Frequency) to encode review text into numeric features for training our model. The performance of three supervised machine learning models Logistic Regression, Support Vector Machine (SVM), and Gradient Boosting have been evaluated by calculating accuracy, precision, recall, F1-score and ROC-AUC. Of these, SVM performed best with an accuracy of 87% and a ROC-AUC of 0.95, suggesting suitability for working with high-dimension text data. The work concludes that SVM on TF-IDF is a strong competitor in fake review detection and there is room for further enhancing with better embeddings or deep learning models.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Basilio, Jorge UNSPECIFIED
Subjects:	P Language and Literature > P Philology. Linguistics > Computational linguistics. Natural language processing 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:	01 Jul 2026 08:26
Last Modified:	01 Jul 2026 08:26
URI:	https://norma.ncirl.ie/id/eprint/9418

Actions (login required)

View Item