Student

Zulfah Athirah

Supervisor

Ainuddin Wahid Abdul Wahab

Collaborator

Tuan Mohd Saud Matt

The development of e-commerce platforms has revolutionized the way people shop online, allowing them to make more proper purchases based on user-generated feedback. However, the authenticity and reliability of these evaluations have become a major worry due to the rising prevalence of phony reviews. This final-year project focuses on creating a fraudulent review detection system for the prominent e-commerce site Shopee.

The goal of this project is to create and deploy an automated method for detecting and categorizing fraudulent reviews on Shopee. The proposed system utilizes state-of-the-art machine learning techniques, including natural language processing (NLP) and sentiment analysis, to analyze textual reviews and extract relevant features. These characteristics are then utilized to build a classification algorithm that can differentiate between real and false reviews.

To achieve this goal, a comprehensive dataset comprising of genuine and fake reviews is collected and annotated for training and evaluation purposes. To capture the semantic and grammatical information inherent in the reviews, various feature extraction approaches such as bag-of-words, TF-IDF, and word embeddings are investigated. Classification technique such as Logistic Regression is used to determine the best successful model for detecting false reviews.

The suggested system is built as a web-based application, with an easy-to-use user interface that allows users to submit Shopee review texts for examination. The system utilizes an efficient and scalable architecture to handle the large volume of incoming reviews and provides real-time feedback on the authenticity of each review.