Year: 2024
Type: dissertation
Abstract: The accumulation of fake reviews are currently widespread among the various e-commerce platforms has created a problem of credibility and distorted purchasing choices. This research’s main objective is to use efficient machine-learning and deep learning approaches to fake review identification. The aim is to classify a given review as fake or real using natural language processing (NLP) techniques and using both the conventional machine learning models along deep learning methods. The dataset contains both real and fake reviews where OLAMA model is used to generate fake reviews and combined with original reviews. To estimate the performance of the proposed model, another test set of 5,000 samples including fake and genuine reviews is used for testing. The machine learning baseline models such as Decision Tree, Random Forest, and Naive Bayes classifiers are used; on the other hand, deep learning models such as LSTM and CNN + BiLSTM are used for comparison purposes. The main goal of the project is to improve the performance of fake review detection which in turn would help improve the overall experience of users of e-commerce platforms.
Source: NORMA
Author: Vasineni, Shiva
Language: English
Topic: Spam and Phishing Detection
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7084208227