Cross-Device Tracking: Systematic Method to Detect and Measure CDT
share
Online advertising, the backbone of the free Web, has transformed the marketing business by creating countless opportunities for advertisers to reach potential customers. The advertising ecosystem exists upon a complex infrastructure composed by intermediate entities and technologies whose main goal is to deliver personalized ads to such customers. In recent years, however, advertisers have started to develop new advanced techniques such as Cross-Device Tracking, to detect and track the user's activity across multiple devices and target them on the all possible device. For this method to work, a variety of user data is collected, aggregated, processed and traded behind the scenes, and often times, without the user-data owner's informed consent. Therefore, despite the enormous value of online advertising and support it offers to the Web, the prevalence and intensity of these tracking and targeting practices prompt serious concerns for the online users' privacy. In this paper, we propose a novel methodology for systematically investigating Cross-Device Tracking and various components affecting its targeting performance. We materialize our methodology into a platform that is able to perform small and large scale automated measurements, under different experimental scenarios, emulated users and settings. By conducting a variety of measurements with our framework, we are able to detect and measure Cross-Device Tracking with average accuracy of 78-96 about its internal mechanisms and its impact on online user's privacy.
READ FULL TEXT