Detection of Fake Grade in Mobile Application Environment



Ranking fraud within the mobile App market refers to deceitful or deceptive activities that have a purpose of bumping up the Apps within the quality list. Indeed, it becomes additional and additional frequent for App developers to use shady means that, like inflating their Apps’ sales or posting phony App ratings, to commit ranking fraud. Whereas the importance of preventing ranking fraud has been wide recognized, there's restricted understanding and analysis during this space. To the current finish, during this paper, we offer a holistic read of ranking fraud and propose a ranking fraud detection system for mobile Apps. Specifically, we have a tendency to initial propose to accurately find the ranking fraud by mining the active periods, particularly leading sessions, of mobile Apps. Such leading sessions may be leveraged for police work the native anomaly rather than world anomaly of App rankings. Moreover, we have a tendency to investigate 3 varieties of evidences, i.e., ranking primarily based evidences, rating {based| based mostly| primarily primarily based} evidences and review based evidences, by modeling Apps’ ranking, rating and review behaviors through statistical hypotheses tests. Additionally, we have a tendency to propose associate improvement primarily based aggregation technique to integrate all the evidences for fraud detection. Finally, we have a tendency to evaluate the projected system with real-world App information collected from the iOS App Store for an extended fundamental quantity. In the experiments, we have a tendency to validate the effectiveness of the projected system, and show the quantify-ability of the detection rule furthermore as some regularity of ranking fraud activities.

Full Text:


Copyright (c) 2016 K. VINAY KUMAR REDDY, C B SUSHMA

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.


All published Articles are Open Access at 

Paper submission: