Classifiers Ensemble for Fake Review Detection
Harish Baraithiya1, R. K. Pateriya2
1Harish Baraithiya, Research Scholar, Department of Computer Science & Engineering, Maulana Azad National Institute of Technology, Bhopal (M.P), India.
2Dr. R. K. Pateriya, Associate Professor, Department of Computer Science & Engineering, Maulana Azad National Institute of Technology, Bhopal (M.P), India.
Manuscript received on 05 February 2019 | Revised Manuscript received on 13 February 2019 | Manuscript published on 28 February 2019 | PP: 730-736 | Volume-8 Issue-4, February 2019 | Retrieval Number: D2650028419/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The growth of e-commerce businesses has attracted many consumers, because they offer a range of products at competitive prices. One thing most purchasers rely on when they purchase online is for product reviews to conclude their decision. Many sellers use the decision to impact the review to hire the paid review authors. These paid review authors target the particular brand, store or product and write reviews to promote or demote them according to the requirements of their hired employees. In view of the effects of these fake reviews, a number of techniques to detect these fake reviews have already been proposed. Because of nature of the reviews it is difficult to classify them using single classifier. Hence in this paper, we proposed an ensemble classifier based approach to detect the fake reviews. The proposed ensemble classifier uses support vector machine (SVM), Naïve Byes classifier and k- nearest neighbor (KNN mutual) classifiers. The proposed technique is evaluated using Yelp and Ott. et al [10] datasets. The evaluation results show that the proposed classifier provides better classification accuracy on both datasets.
Keyword: Fake Review Detection, Ensemble Classifiers, Behavioral Analysis, Opinion Spam.
Scope of the Article: Predictive Analysis