Trajectory-User Linking Is Easier Than You Think
This work challenges the assumption that complex deep learning is necessary for TUL, potentially simplifying applications like personalized recommendations and criminal detection, though it is incremental in demonstrating the sufficiency of simple methods.
The paper tackles the Trajectory-User Linking (TUL) problem by showing that simple heuristics based on raw visit patterns, such as using a single check-in per trajectory, can achieve up to 85% accuracy in user identification, and scales to over 100k users, a three-order-of-magnitude increase over state-of-the-art methods.
Trajectory-User Linking (TUL) is a relatively new mobility classification task in which anonymous trajectories are linked to the users who generated them. With applications ranging from personalized recommendations to criminal activity detection, TUL has received increasing attention over the past five years. While research has focused mainly on learning deep representations that capture complex spatio-temporal mobility patterns unique to individual users, we demonstrate that visit patterns are highly unique among users and thus simple heuristics applied directly to the raw data are sufficient to solve TUL. More specifically, we demonstrate that a single check-in per trajectory is enough to correctly predict the identity of the user up to 85% of the time. Moreover, by using a non-parametric classifier, we scale up TUL to over 100k users which is an increase over state-of-the-art by three orders of magnitude. Extensive empirical analysis on four real-world datasets (Brightkite, Foursquare, Gowalla and Weeplaces) compares our findings to state-of-the-art results, and more importantly validates our claim that TUL is easier than commonly believed.