The School of Computer Science at the University of Windsor is pleased to present …
Spam Review Detection Using Similarity Representation
MSc Thesis Proposal by:
Ashraf Neisari
Date: Tuesday, February 26th, 2019
Time: 1: 00 pm – 2:00 pm
Location: Lambton Tower, Room 3105
Abstract:
Online public reviews have significant influence on customers, while they can also can manipulate or poison the public opinion over a subject. They have profound impact on businesses through promoting or demoting their products, services and reputation. Posting fake (spam) reviews is done by an individual or a group of people called “spammers”. As such, there is a need to develop effective techniques to distinguish spam reviews from the genuine ones and prevent spam review propagation.
Opinion mining and spam review detection has gained significant importance for many organizations and companies. Researchers have used many different approaches during the past few of years to detect spam reviews and spammers, including conventional machine learning and deep learning methods.
In this thesis, we propose to develop a fast method to detect fake reviews utilizing natural language processing (NLP), including Google’s word2vec approach, shallow and deep machine learning, incorporating similarity representation of reviews. Well-known datasets will be used to benchmark our proposed method.
Thesis Committee:
Internal Reader: Dr. Arunita Jaekel
External Reader: Dr. Maher Abdelkalek
Advisor: Dr. Luis Rueda
Co-Supervisor: Dr. Sherif Saad Ahmed
Thesis Proposal Announcement
5115 Lambton Tower
401 Sunset Avenue
Windsor, Ontario
N9B 3P4
(519) 253-3000 Ext. 3716