Implementation of Novel Machine Learning Methods for Analysis and Detection of Fake Reviews in Social Media

Abstract

Now-a-days social media plays a key role in our daily life. For many things we rely on this social media. This social media plays a very crucial role in several fields. People started believing the things which are written in internet before taking their decisions, such as having a look on reviews which are written in the social media for various purposes like buying a product in online or booking a hotel room for vacation or for visiting a place, for all these things people rely on these reviews which are written by several people. By these reviews there will be advantages as well as disadvantages. Some people to improve their company standards or for high lightening their products, they generate few fake reviews which attracts the users towards them and by that people starts choosing them. A target object’s positive ratings may draw more consumers and boost sales, whereas a target object’s negative reviews may result in less demand and lower sales. Our research aims to determine if a review is genuine or fraudulent. To avoid these fake reviews and to go with genuine reviews we need a fake review detection system. In this system some machine learning techniques are used. The proposed method consists of machine learning algorithms. They performed a benchmark analysis with different types of (1) traditional ML algorithms like logistic regression (LR), support vector machines (SVM), decision trees (DT), Naive bayes (NB), random forests (RF), & XG Boost (XGB), as well as an ensemble learning approach of such algorithms, and (2) cutting-edge ML algorithms like bidirectional long short-term memory (BIIS TM), etc. These algorithms assist in identifying bogus reviews. These algorithms are all contrasted with one another to provide precise results.