ABSTRACT As users increasingly interact with AIoT systems through multiple Internet‐connected devices, conventional single‐device tracking fails to capture distributed user behaviours, limiting the effectiveness of personalization and targeted advertising. Cross‐device tracking therefore plays a crucial role in constructing unified user profiles from heterogeneous device traces. This paper proposes a novel cross‐device tracking approach that links users' mobile phones and tablet computers. To the best of our knowledge, this is the first study focusing specifically on these two device types. Our key insight is that users exhibit consistent operational characteristics across devices, despite differences in size and weight. We collect data from five motion sensors during four common device operations and design a multi‐scale spatio‐temporal twin network to capture behavioural similarities. Experimental results demonstrate that the proposed approach achieves an accuracy of 77.8% in a 18 participants scenario. We further analyse the impact of user operation types, model components and sensor data types on tracking performance. This work provides an effective solution for phone–tablet cross‐device tracking and offers practical support for future advertising technologies.
Guo et al. (Thu,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: