Decision Tree Analysis for Identifying False Apps in the Google Play Store

Abstract

Tablets, smartwatches, and lightweight laptops are all becoming more commonplace. When it comes to mobile app stores, Android is king. Because Android is freely available, hackers often attack mobile applications. The fact that most people have never downloaded or used an app before does not help things. The execution of the mobile app in issue and the user's cooperation were prerequisites for all previous methods of identifying dangerous or fraudulent applications. A system for scanning Google Play for malicious software and unearthing cheating developers that manipulate their search rankings. Combining data from rankings, peer reviews, and ratings may help expose a malicious program. Finally, we can attain a high degree of accuracy in distinguishing harmful, fraudulent, and genuine apps within typical datasets by merging the behaviors of all foreground programs. We employ incremental learning to explain more and more data sets. It worked effectively in tandem with other anti-fraud data. For ranking fraud to be detected, it is important to mine the most popular sessions of mobile applications. I.