The ever-increasing usage of mobile phones challenges carriers to evolve and improve their services based on their users' behavior. Nevertheless, it remains uncommon the usage of data about human mobility for discovering patterns which could support the dynamic resource allocation in urban areas. In particular, occasional large scale events impose patterns that may strongly differ from those observed in daily routines. In this paper, we present a methodology to characterize the workload dynamics of a mobile phone network during large scale events. Our study is based on calls made by mobile phone users of a major Brazilian cellphone carrier during soccer matches with a large number of attendees at Rio de Janeiro's urban area. The results of applying the proposed methodology may be used to improve the understanding on how users move within an urban area towards the location of a large scale event and how they disperse afterward, helping the management of mobile phone networks. Moreover, the uncovered mobility patterns may drive the development of target applications and gadgets for large scale events, such as World Soccer Cup matches.
Recently it has been observed a world wide increase of online sales, mainly due to agility to buy and attractive prices that are offered on the Web. However, fraud has also been increased on the same rate or more. In order to address this problem it is very important to understand the characteristics of fraudsters and their typical behavior. On the tourism e-market it is not different, thus millions of frauds occur each year. In this work we analyze a representative amount (thousands) of online transactions of a tourism Web system. We try to understand the characteristics of fraudsters with the main goal to support decision of e-payment evaluation of transactions. Our results are promising, achieving up to 64% of increase in accuracy in comparison to the baseline.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.