International Research Journal of Engineering and Technology (IRJET) Volume: 04 Issue: 02 | Feb -2017
www.irjet.net
e-ISSN: 2395 -0056 p-ISSN: 2395-0072
A Review : Mobile App Recommendation Based On Rating Review & Ranking Dharti A. Bobade1, Prof.V.S Gangwani2 1M.E.Student, 2Assistant
Department of Computer Science & Engineering, H.V.P.V College of Engineering, Amravati, India Professor, Department of Computer Science & Engineering H.V.P.V College of Engineering, Amravati, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - The Mobile App is a very popular and well known concept due to the rapid advancement in the mobile technology. Due to the large number of mobile Apps, ranking fraud is the key challenge in front of the mobile App market. There are millions of apps are available in market for the application of mobile users. However, all the mobile users first prefer high ranked apps when downloading it. To download application smart phone user has to visit play store such as Google Play Store, Apples store etc. When user visit play store then he is able to see the various application lists. This list is built on the basis of promotion or advertisement. User doesn’t have knowledge about the application (i.e. which applications are useful or useless). So user looks at the list and downloads the applications. But sometimes it happens that the downloaded application won’t work or not useful. That means it is fraud in mobile application list. To avoid this fraud, we are making application in which we are going to list the applications. In this paper, we provide a brief view of ranking fraud and propose a ranking fraud detection system for mobile Apps. Specifically, we first propose to accurately locate the ranking fraud by mining the active periods by using mining leading session algorithm. Furthermore, we investigate three types of evidences, i.e., ranking based evidences, rating based evidences and review based evidences, by studying historical records. we used an optimal aggregation method to integrate all the evidences for fraud detection. Finally, we evaluate the proposed system with real-world App data collected from the Google App Store for a long time period. In the experiments, we validate the effectiveness of the proposed system, and show the scalability of the detection algorithm as well as some regularity of ranking fraud activities. Key Words: Mobile Apps, ranking fraud detection, historical ranking records, evidence aggregation, review, ranking and rating.
© 2017, IRJET
|
Impact Factor value: 5.181
|
1.INTRODUCTION AS smartphones emerges new technologies like android and iOS operating system took a boost in market. Mobile application started growing at such a high rate. As a study says millions of apps are there on apple’s app store and on Google Play. This started a new business in computer world and became a reason to earn thousands of dollars and downloads. Daily leaderboard is published by these markets contains the most popular apps which will consequently be downloaded and rated most high by users. Some developers may use some marketing strategies like an advertisement campaign for promotion of their app. However this part of technology is also not safe from threats. Mobile app market, we refer it as market, is manipulated by some fraudulent app developers to bump up their app high in the rank list, as an app in leaderboard confirms high downloads and high income. Shady means are used to make such a fraud and implemented using “bot farms” which is also called “Human water armies”. In this area some literature survey is there, for example, spam detection for web ranking, mobile app recommendations, and some online review based spam detection. Our study thus focuses on an integrated approach, for various evidences, to find Mobile App ranking fraud and also recommend the most relevant App that is most genuine. For this we have to go through challenges like first we need to find at what time the fraud is happening it means exact time of fraud is needed. Secondly we know that there is tremendous number of Apps present in market so it is nearly impossible to physically mark ranking fraud for every App, so it’s crucial to automatically distinguish fraud without utilizing any essential data. Mobile Apps are not commonly ranked high in the leader board, but instead just in a few events ranking frauds more often than not happens in leading sessions. In this way, fundamental target is to recognize ranking fraud of mobile Apps inside of leading sessions. Initially propose an efficient algorithm to recognize the main sessions of every App depends on its previous ranking records. By then, with the examination of Apps’ ranking practices, find the fake Apps consistently have unique ranking examples in every leading session contrasted with ordinary Apps. Along these lines, some fraud confirmations are portrayed from Apps’ previous ranking
ISO 9001:2008 Certified Journal
|
Page 391