مقاله انگلیسی رایگان در مورد الگوریتم حفظ حریم خصوصی مبنی بر مسیر مکرر – الزویر 2018

With the rapid development of location-based services, many mobile positioning devices have emerged, such as car navigation, GPS-enabled mobile phones, tablet PCs and position sensors. As a result, major manufacturers have launched their own location-based service applications, with which users can send their location and query content to the location server, and then, the location server will return the corresponding query results—for example, Google Maps for navigating services when travelling, Baidu glutinous rice for geo-location search services for nearby restaurants, and WeChat for social services that can share geo-labelling. These applications can be divided into two types. One is online applications based on the real-time location provided by the user, which require the corresponding services—e.g., locationbased services (LBS), push services based on geo-real-time information and real-time monitoring of moving objects—to be provided. The other is offline applications, in which location service providers or other agencies collect and analyse mobile data or publish the trajectory data to third parties. For example, through the excavation and analysis of trajectory data, it is possible to optimize traffic network and traffic management strategies and analyse user behaviour to support business decisions. Although these two types of applications have brought great convenience to people’s lives [1], disclosure of their private locations to potentially untrusted LBS service providers poses privacy concerns. Two surveys reported in July 2010 showed that more than half of users who use LBS services are concerned about the disclosure of location privacy [2], and 50 percent of U.S. residents who have a profile on a social networking site are concerned about their privacy [3]. The research results confirmed that location privacy is one of the key obstacles to the success of location-dependent services [4]. Privacy in offline applications is more challenging than online applications because an attacker can infer the user’s location information by using the spatial and temporal correlations in the user’s location samples. However, the trajectory formation is important for many applications in real life, such as business analysis, city planning, or transportation planning. Therefore, privacy protection in offline applications and trajectory data publication has increasingly drawn attention from the industry and academia. Many approaches have been proposed for preserving privacy in trajectory data publishing, but most do not consider the usability of data for publishing. The result is that our privacy may be preserved well while the trajectory data are of no value to the applications (e.g., city planning), leading to a significant loss of information. Since most publishing information is used for data mining and analysis, it is necessary to focus on the hot region or frequent path. In this paper, we address this problem by using frequent path to preserve data privacy, which has not been studied in previous work, so that not only can data privacy be preserved but also the usability of data can be increased.