M-Atlas 2.0: the DMQL. The new data mining query language

Romano V., Trasarti R., Ghelli G.

M-Atlas  Database Applications 

In the last years the researchers at KDD Lab have used the M-Atlas system for their mobility data analyses and contributed to expand the M-Atlas functions and features. After so much time, the original design of M-Atlas needed to be reviewed in order to understand how it could be adapted to improve the performances and ease the work of the researchers performing analyses. In the last months the M-Atlas Core has been almost completely redesigned in the way it connects to the PostgreSQL database and how datasets and software methods are managed. The new design of the software methods as Plug-ins makes it very simple to invoke a method. Now the old DMQL module has to be modified in order to interact with the new Core and to take advantage from the new design. Moreover, as the DMQL language can be used by data scientists to directly write DMQL code or to interact through the GUI or the Web Service which will generate DMQL code, we built a new intuitive language that was easy to learn and similar to other known languages.

Source: ISTI Technical reports, 2016