Fake Comments Detection with Sentiment Anatomy using Iterative Sequential Minimal Optimization Algorithm
Vanshita Pansari1, Amit Kumar Manjhvar2

1Vanshita Pansari*, Department of Computer Science and Engineering, Madhav Institute of Technology and Science, Gwalior, India.
2Amit Kumar Manjhvar, Department of Computer Science and Engineering, Samrat Ashok Technological Institute, Vidisha, India.
Manuscript received on May 08, 2020. | Revised Manuscript received on May 19, 2020. | Manuscript published on June 10, 2020. | PP: 41-46 | Volume-9 Issue-8, June 2020. | Retrieval Number: 100.1/ijitee.H6137069820 | DOI: 10.35940/ijitee.H6137.069820
Open Access | Ethics and Policies | Cite | Mendeley
© 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: It is significant to create electronicon stream markets, on stream communication networks, peer-to-peer functions, social media providerson stream and convenience customers. In reality, web based amenities are specially designed to overcome the risk of uncertainties & distrust inherent in the main concern of ecommerce applications & to increase the robustness of the system& resistance against fake clients & unbelievers. The aim of the Ecommerce platform is, moreover, to embrace one of the most efficient methods for understanding and evaluating user attempts to expose fraudsters. Or else, the fundamental objective of ecommerce amenities to exploit the profit & purchase rate, will be endangered & deteriorated through fake and ill-intentioned users. Individuals and organizations need to detect fake Comments. With disappointing and hidden features, it is difficult to identify counterfeit Comments simply by looking at a single Comments text. It is also why it is a difficult task to identify falsified Comments. This paper uses the sentiment anatomy (SA) tool for the identification of fake Comments to analyzeon stream film Comments. The texts and the SA system are used for a specific dataset of film Comments. We particularly compared the supervised SVM & SMO machine-learning process with the feeling classification methods of the analyzes in two different cases, without stopping phrases. Measured outcomes display that SMO process compared to the SVM process for both methodes, &it arrives at the maximum precision not only in the classification of text but also for finding duplicate analyses. 
Keywords: Data Mining, Reputation System, on stream Movie Comments, Machine Learning method, SMO, SVM, Sentimental Anatomy, Text Classification, Fake Comments Detection.
Scope of the Article: Data Mining