دانلود رایگان مقاله انگلیسی الگوریتم حفظ حریم خصوصی جدید مبتنی بر مسیر تکرار شونده برای انتشار داده های مسیریابی - نشریه الزویر

Abstract

Existing location-based services have collected a large amount of location data, which contain users’ personal information and has serious personal privacy leakage threats. Therefore, the preservation of individual privacy when publishing data is receiving increasing attention. Most existing methods of preserving user privacy suffer a serious loss in data usability, resulting in low usability of data. In this paper, we address this problem and present TOPF, a novel approach for preserving privacy in trajectory data publishing based on frequent path. TOPF aims to achieve better quality of trajectory data for publishing and strike a balance between the conflicting goals of data usability and data privacy. To the best of our knowledge, this is the first paper that uses frequent path to preserve data privacy. First, infrequent roads in each trajectory are removed, and a new way is adopted to divide trajectories into candidate groups. A new method for finding the most frequent path is then proposed, and then, the representative trajectory is selected to represent all trajectories within a group. Experimental results show that our algorithm not only effectively guarantees the privacy of the user but also ensures the high usability of the data.

6. CONCLUSION

It is becoming increasingly important to preserve individual privacy when publishing trajectory data. How to widely use the data without violating user location privacy concerns has attracted the attention of scholars.

Most previous works examining the preservation of privacy in trajectory data publishing suffer a serious loss in data usability. We address this problem by focusing on the high usability of data and present TOPF, a novel approach for preserving privacy in trajectory data publishing that uses frequent path to strike a balance between the conflicting goals of data usability and data privacy.

TOPF not only can provide data privacy protection in data publishing but also can ensure the high usability of the data. This is because TOPF first removes all infrequent roads, which avoids identity linkage and guarantees that each road has k-anonymity. TOPF then adopts a new way to divide trajectories into candidate groups and proposes a new method for finding the most frequent path, and then, we select a representative trajectory to represent all trajectories within the group. We evaluate TOPF in terms of average error rate, standard deviation and F-measure values, and we conduct an experiment comparing TOPF with ICBA, Prefix and NWA. The result demonstrates that our approach is superior to the others.

 In future work, the proposed method can be extended in the following two aspects: (1) how to update the parameters θ, δ in this article to obtain a better result; (2) how to optimize the grouping method to reduce the error rate and preserve a more frequent pattern. Moreover, how to make this algorithm more efficient is also worth considering.