Abstract
Acknowledgements
Table of Contents
List of Tables
List of Figures
Introduction
Online Reviews: Electronic Marketing
The Problem Of Fake Reviews
Existing Approaches
Aims and Objectives
Related Work
Overview
Background Information
Text Based Features
Ensembles
Research Methodology
Overview
Research Questions
Data Collection and Pre-processing
Feature Engineering and Data Analysis
Use of Individual Feature Sets
Use of Combined Feature Sets
Ensemblement of Classifiers
Evaluation Metrics
Tools and Datasets Used
Feature Engineering
Overview
Stylometry
Readability
Syntactic Analysis
Sentiment Analysis
Lexical Semantics
Personality and Tone Analysis
Summary
Results with Individual Feature Sets
Overview
Individual Classifiers
Individual Classifiers and Ensemble Methods
Conclusions
Results with Combined Feature Sets
Overview
Individual Classifiers
Individual Classifiers and Ensemble Methods
Conclusions
Classifier Ensemblement
Overview
Classifiers that Use Only Individual Feature Sets
Classifier Selection Schemes
Classifiers That Use Only Combined Feature Sets
More Complex Ensembles
Conclusions
Accuracy And Diversity
Overview
The Winnowing Process
The Dynamics Between Accuracy And Diversity
Conclusions
Discussion
Overview
Data
Methodology
Experiments
Results
Conclusion
Main Findings
Suggestions for Further Work
Bibliography