An Implementation to Detect Fraud App Using Fuzzy Logic

Abstract

Fraudulent behavior is most popular in app stores like Google play store, Apple�s app store, etc. The popularity information in app stores, such as chart rankings, user ratings, and user reviews, provides an extraordinary chance to recognize user experiences with mobile apps. Many fraud app detection tools are available these days which extract evidences of reviews and ratings to detect the fake apps with different approaches. But most of the existing tools work on two groups i.e., good and bad. So, we propose a system that works on more than two groups namely, very bad, bad, neutral, good and very good. Each group has been assigned a score which will improve the differentiation of reviews and ratings. For this the proposed system uses fuzzy logic algorithm. We have performed experimentation on 80 app ids taken from App-Review-Dataset, results show that proposed method is efficient in terms of accuracy and time required for retrieval.