Learning Fraud Rating for Mobiles using Aggregation Appliance

Objective: Identification of ranking frauds in mobile application is one of the interesting issues in the current research field and technology. Fake ranking can be maintained by the various application developers with programs like "bot farms" or "human water armies". Methods/Analysis: In this paper we are proposing a novel approach for rank fraud detection of mobile application based on improved leading sessions with session duration, rating based, review based and a novel aggregation mechanism. Findings: Efficient rank implementation with session identification and removal of fake injection of comments over products. Improvement: Our proposed approach gives more efficient results than the traditional approaches.