Playing Hide and Seek with Mobile Dating Applications

Abstract

Abstract. Recently, a wide range of dating applications has emerged for users of smart mobile devices. Besides allowing people to socialize with others who share the same interests, these applications use the location services of these devices to provide localized mapping of users. A user is given an approximation of his proximity to other users, making the application more attractive by increasing the chances of local interactions. While many applications provide an obfuscated location of the user, several others prefer to provide quantifiable results.This paper illustrates that the user's location can be disclosed, with various degree of approximation, despite the obfuscation attempts. Experimenting with four of these applications, namely MoMo, WeChat, SKOUT and Plenty of Fish, we show that an attacker can easily bypass the fuzziness of the results provided, resulting in the full disclosure of a victim's location, whenever it is connected.

Security and Trust Management

“…This complexity is somehow intrisic to the problem when assuming that the purpose of an action is determined by its position in a workflow, as we do in this paper. As a consequence, achieving a purpose amounts to the succesful execution of the associated workflow, which is called the Workflow Satisfiability Problem (wsp) in the literature [12] and is known to be NP-hard already in the presence of one sod constraint [34]. Several works have proposed techniques to solve both the off-line [27,34] and the on-line [6,8] version of the wsp but none considers purpose-aware policies as we do here.…”

“…This is so because-as discussed above-a purpose is associated to a workflow so that its successful execution implies the achievement of the purpose. The problem of checking (offline) if a workflow can successfully terminate, known as the Workflow Satisfiability Problem (wsp), is already computationally expensive with one sod [34], and moreover, the on-line monitoring of authorization constraints requires to solve several instances of the wsp [12]. For purpose-aware policies, this implies that it is necessary to solve an instance of the wsp per user request of executing a task in the workflow associated to a given purpose.…”

“…The wsp is known to be a computationally expensive activity; it is already NP-hard with one sod constraint [34]. To make things even more complex, at run-time, we need to solve several instances of the wsp, one per user request of executing a task in a workflow associated to a purpose.…”

Lightweight private proximity testing for geospatial social networks

“…by displaying an estimation of the distance of a user from others. As it has been shown independently in [11,12] the provided obfuscation mechanisms can be trivially bypassed, exposing the exact location of the users. The exposure of a user's location can disclose a lot of sensitive information.…”

InnerCircle: A parallelizable decentralized privacy-preserving location proximity protocol

“…Deemed as a serious privacy violation, the app had since been banned from using the Foursquare API and removed from the app store. Recent research systematizes these attacks and identifies a number of LBS where it is possible to reveal the user' location even if some of the LBS approximate and obfuscate distance information [27], [22].…”