Tour M-Atlas


Thanks to a Flock discovering algorithm M-Atlas is able to capture citizens which move in together for a certain amount of time. This can be used to spot traffic jams in trajectory car data understating the critic areas of a city.

Trajectory Patterns

M-Atlas integrates the Trajectory Pattern algorithm which is able to extract the frequent spatio-temporal patterns over a trajectory dataset. They are represented in terms of sequence of regions with temporal intervals frequently used between each pair. This can be used to detect local flows of citizens moving over the city.


A particular kind of statistical analyses are the space distribution where a measure is plotted on a map to represent a distribution such as density, average speed or CO2 impact.


M-Atlas is able to analyze the mobility in terms of interconnection between adjacent areas in orders to extract the real borders between less string connected regions and less connected ones.

Mixing tools and analyses

M-Atlas and DMQL gives the possibility of mixing the analyses discovering more useful and detailed knowledge from the trajectory data. An example is to classify the trajectories into systematic or occasional looking at the personal mobility of each user and then use this classification to evaluate the systematic behavior impact over the access path in a city.

Personal Mobility

With M-Atlas is possible to study the global behaviors but also the personal mobility of the user in order to modeling it such as extracting mobility profiles. Thanks to the DMQL the process applied for one user may be replicated over all of them systematically.

Trajectory Clustering

M-Atlas integrates a powerful clustering algorithm (density based) equipped with several distance functions able to group the trajectories in different ways adapting to the user needs. An example is the discovery of the access paths of the city. This is another brick that can be used in a more complex analytical process.