ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Effective Evidence Aggregation to Identify the Ranking Fraud for Mobile APPS Using KNN

Journal: International Journal of Scientific Engineering and Research (IJSER) (Vol.4, No. 11)

Publication Date:

Authors : ; ;

Page : 148-154

Keywords : Mobile Apps; Evidence Aggregation; Positioning misrepresentation; Classification; KNN;

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

Nowadays everyone having smart phone. We have to install so many applications in that smart phone.Presently, the advancements made in the mobile technology are the production of mobile apps. Due to the more number of mobile apps, there is a chance of fraud mobile applications is of greater existence. Ranking fraud is the key challenges persist in the mobile app market. It defines the report over the apps in order to place them in leader board. In this study, we consider the fraud activities in the mobile apps, as a significant one. We propose ranking scheme for recognizing the fraudulent apps.While the essentials of envisioning situating deception have been for the most part seen, there is obliged understanding and research here. To this end, in this paper, we dedicate an all-encompassing view of positioning extortion and propose a positioning misrepresentation location framework for versatile Apps, We first propose to just determine the positioning misrepresentation by mining the dynamic time frames, to be specific driving sessions, of versatile Apps. Such driving sessions can be used for identifying the neighbourhood irregularity rather than worldwide inconsistency of App rankings. All, we examine three sorts of copies, i.e., positioning based confirmations, evaluation based proofs and audit based proofs, by displaying Apps' positioning, rating and survey practices through measurable theory tests. What's more, we offer an enhancement based conglomeration strategy to incorporate every one of the confirmations for extortion recognition. Specifically, it is proposed to exactly discover the mining to posture blackmail the dynamic time frames, to be particular driving sessions, of compact Apps. The KNN algorithm is applied to enhance effectiveness and precision of the application, we approve the sufficiency of the proposed framework and demonstrate the versatility of the identification, calculation and some normality of positioning misrepresentation exercises.

Last modified: 2021-07-08 15:44:11