GraphTrack: A Graph-Based Mostly Cross-Device Tracking Framework

From Beasts of Bermuda
Jump to: navigation, search


Cross-device monitoring has drawn rising consideration from each commercial companies and the general public because of its privateness implications and functions for consumer profiling, personalized providers, etc. One explicit, vast-used type of cross-device tracking is to leverage browsing histories of user units, e.g., characterized by a list of IP addresses used by the gadgets and ItagPro domains visited by the gadgets. However, current shopping history based strategies have three drawbacks. First, they can't seize latent correlations amongst IPs and domains. Second, their efficiency degrades significantly when labeled machine pairs are unavailable. Lastly, they are not strong to uncertainties in linking looking histories to gadgets. We suggest GraphTrack, iTagPro website a graph-based mostly cross-gadget tracking framework, travel security tracker to trace users throughout totally different gadgets by correlating their browsing histories. Specifically, we suggest to mannequin the complicated interplays amongst IPs, domains, iTagPro website and devices as graphs and seize the latent correlations between IPs and between domains. We construct graphs that are robust to uncertainties in linking browsing histories to devices.



Moreover, ItagPro we adapt random stroll with restart to compute similarity scores between units primarily based on the graphs. GraphTrack leverages the similarity scores to perform cross-device monitoring. GraphTrack does not require labeled gadget pairs and can incorporate them if obtainable. We evaluate GraphTrack on two actual-world datasets, i.e., a publicly accessible mobile-desktop tracking dataset (round a hundred users) and a multiple-gadget monitoring dataset (154K users) we collected. Our results show that GraphTrack considerably outperforms the state-of-the-artwork on both datasets. ACM Reference Format: Binghui Wang, Tianchen Zhou, Song Li, Yinzhi Cao, Neil Gong. 2022. GraphTrack: A Graph-primarily based Cross-Device Tracking Framework. In Proceedings of the 2022 ACM Asia Conference on Computer and Communications Security (ASIA CCS ’22), May 30-June 3, 2022, Nagasaki, Japan. ACM, New York, NY, USA, 15 pages. Cross-machine tracking-a way used to identify whether or not varied gadgets, equivalent to mobile phones and desktops, have widespread homeowners-has drawn much consideration of each commercial corporations and most people. For example, Drawbridge (dra, 2017), an promoting firm, goes beyond traditional machine monitoring to determine gadgets belonging to the same person.



Because of the rising demand for cross-system tracking and corresponding privateness issues, the U.S. Federal Trade Commission hosted a workshop (Commission, 2015) in 2015 and iTagPro website launched a staff report (Commission, 2017) about cross-device monitoring and trade rules in early 2017. The growing interest in cross-device tracking is highlighted by the privateness implications associated with monitoring and the functions of tracking for user profiling, personalized services, and user authentication. For example, a bank application can adopt cross-system tracking as part of multi-factor authentication to increase account security. Generally speaking, cross-device monitoring mainly leverages cross-machine IDs, background environment, or searching historical past of the units. As an illustration, iTagPro USA cross-machine IDs might include a user’s e-mail handle or username, which aren't relevant when users don't register accounts or do not login. Background environment (e.g., ultrasound (Mavroudis et al., 2017)) additionally can't be utilized when units are used in several environments akin to residence and workplace.



Specifically, shopping historical past primarily based tracking makes use of source and destination pairs-e.g., the client IP address and the vacation spot iTagPro website’s domain-of users’ searching information to correlate completely different devices of the identical consumer. Several searching history primarily based cross-gadget monitoring strategies (Cao et al., 2015; Zimmeck et al., 2017; Malloy et al., 2017) have been proposed. For instance, IPFootprint (Cao et al., 2015) uses supervised learning to investigate the IPs generally used by gadgets. Zimmeck et al. (Zimmeck et al., 2017) proposed a supervised method that achieves state-of-the-artwork performance. Specifically, their methodology computes a similarity score by way of Bhattacharyya coefficient (Wang and Pu, 2013) for a pair of gadgets based mostly on the common IPs and/or domains visited by both units. Then, they use the similarity scores to track units. We name the strategy BAT-SU because it makes use of the Bhattacharyya coefficient, where the suffix "-SU" signifies that the strategy is supervised. DeviceGraph (Malloy et al., 2017) is an unsupervised method that fashions devices as a graph based mostly on their IP colocations (an edge is created between two units in the event that they used the identical IP) and applies community detection for tracking, i.e., the gadgets in a group of the graph belong to a consumer.