Please use this identifier to cite or link to this item:
Title: Spam review detection (II)
Authors: Tussupbekov, Yerken.
Keywords: DRNTU::Engineering::Computer science and engineering::Data::Coding and information theory
Issue Date: 2013
Abstract: Online opinions have become an essential part of decision making for millions of web users. However, in a pursuit of profit or success, imposters try to deceive people by opinion spamming to promote or demote a certain targets. The seriousness of the problem has attracted significant attention from different parties. Big companies develop new approaches of enhancing the filtering systems to detect spam, while spammers come up with new ways to disguise themselves. In this paper, we study two major approaches for spam detection based on linguistic and behavioral features. While most of the past researches focused on either of the methods, in this work we combine the two together in an attempt to find the optimal spam filtering approach. We will take supervised learning approach, as the new ways of detecting training spam will be proposed and put to the test. An in-depth investigation explores new principles for dataset construction that allows us to develop a classifier reaching remarkable 83.4% accuracy in spam filtering. Furthermore, additional enhancement to the developed system will be proposed, that could help to achieve even better performance.
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
Spam Review Detection (1).pdf
  Restricted Access
Spam Review Detection19.56 MBAdobe PDFView/Open

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.