International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169 Volume: 3 Issue: 12 6590 - 6592 ________________________________________________________________________________________________________ 6590 IJRITCC | December 2015, Available @ http://www.ijritcc.org ________________________________________________________________________________________________________ Determination of Ranking Fraud for Mobile Applications Sadigale Rohit Birudev 1 Department of computer engineering, DGOI, FOE, Daund. Pune, Maharashtra- India sadigalerohit@gmail.com Amrit Priyadarshi 2 Department of computer engineering, DGOI, FOE, Daund. Pune, Maharashtra- India amritpriyadarshi@gmail.com Abstract: Mobile application is important for all the smart phone users to play or perform different tasks .There large numbers of mobile application developers are available; they can develop the different mobile applications. For making lager users for their mobile applications some developers refers fraudulent activities. Due to these fraudulent activities the mobile applications jump up in the application popularity list. Such fraudulent activities are used by more and more application developers. The fraudulent activities are like mobile application rating, review and its ranking. For this issue large number of users makes a mistake and downloads the mobile applications which have higher review, rating and ranking. So in this paper, we determine the ranking fraud happens in mobile applications and develop ranking fraud detection system. For identifying the ranking fraud, first we consider leading sessions of mobile applications. Then we examine three types of evidences, these are 1) Ranking based evidence, 2) Rating based evidence and 3) Review based evidence. After this we can aggregate all these evidences for fraud detection. Finally, we develop a system that determines fraud happened in mobile applications. Keywords: Mobile Applications, ranking fraud detection, ranking evidence, rating evidence, review evidences. __________________________________________________*****_________________________________________________ I. Introduction: There are many mobile applications that have grown at very higher rate over past few years. For example: - at the end of April 2015 there were more than 1.9 million applications at Apple’s application store and Google play store. For provoking the development of applications, many application stores provided daily application leader boards, which gave a chart of ranking of most popular applications. For promoting the mobile applications the application leader board is one of the most important ways. The application which has the higher rank on the leader board usually gives the huge number of downloads and million dollars in revenue. Hence, to have their applications ranked as high as possible in such application leader board the application developers does many advertisements to promote their applications in order to get the higher downloads and revenue. But, instead of being dependent such traditional marketing solutions, the fake application developers used some fraudulent techniques to boost their applications and also manipulate the charts rankings on an application store. This is done by using “bot farms” or “human water armies” to inflate the application downloads, ratings and reviews in a very small time. For example, an article from VentureBeat [1] reported that, when an application was launched with the help of ranking manipulation, it could be jumped from number 1,800 to the top 25 in Apple’s top free leader board and larger than 50,000-100,000 new users could be acquired within a couple of days. In fact, such ranking fraud raises great concerns to the mobile application industry. II. Related Work: The related works of this study consist of spam detection, which contains the web ranking spam refers to rank the selected web pages which belongs to the web site that we have to rank. As there are large numbers of web sites are available, so each developer of website wants higher rank of their website. That fraudulent website ranking spam detection is encountered. Another task is related to the online review spam. When the users buy any product or applications they can give their review, so that review may be fraudulent by the developer. Such Fraud review spams are detected, and the latest fraud is related to the mobile applications. III. Methodology: To highlight such problem, in this paper, we have to develop a ranking fraud detection system for mobile applications. To cover such a problem we have to decide the ranking fraud. As the ranking fraud not always happen in the whole life duration of a mobile application, so we need to detect the time when fraud happens. In the real world, the huge numbers of mobile applications are developed by the developer; and that why it is difficult to manually detect ranking fraud for each application. Hence it is important to find out the different way that automatically detects ranking fraud without considering any of the benchmark information. The leather board which is chart that display ranking of mobile applications, and due to the dynamic nature of the chart rankings, it is too difficult to identify and