FAKE REVIEW DETECTION ON GOOGLE PLAY STORE APP

Abstract

There are millions of apps in google play store all of which allows the users to write reviews about the usage, quality, performance, issues on the particular app. These reviews help new users to get a brief insight into the app and help developers to improve constantly. Positive reviews help build the reputation of the app while negative reviews degrade the same. Now a days there are a lot of fake reviews posted on the google play store both of positive and negative nature which gets developers into trouble and the users not to see the reality about these apps. Fake review detection helps weed out these fake reviews and let the users and developers view an actual image of the app as it is. Previous methods of fake review detection for app store reviews lack on taking into consideration the important features necessary for high accuracy. In this paper we propose a supervised model for fake review detection of google play store apps which uses both review centric features and reviewer centric features and based on these features we build a naïve bayes classifier which successfully detects fake reviews on a given dataset of app reviews.