Artificial Dataset: An artificial dataset which is based on the transformation of a seed semantic trajectory according to different criteria and rates. Over that seed trajectory, we generate a set of η transformed semantic trajectories in different transformation rates SrT = {A’1 , . . . , A’η } where r is the rate and T is the type of transformation. Five types of transformation were applied over the seed trajectory in the experimental study. The seed trajectory was created with 10 disjoint stops (i.e. they do not match in any dimension with each other), and for each transformation, we used the parameters η = 50 representing the number of generated transformed semantic trajectories, and varying values for the transformation rate r. Although the number of stops in semantic trajectories can vary according to different domains the number of elements in a semantic trajectory is in a different order of magnitude when compared to raw trajectories, i.e., it has a significantly lower number of elements. For that reason, an initial number of 10 stops in the seed trajectory is appropriate, even though considering that the stops in the same trajectory are disjoint, identical results would be found with more stops per trajectory (e.g., 50 stops).