Scalable reasoning over big and heterogeneous data

Data comes from many different sources, and is typically visualised using temporal or spatial representation. When combined, correlated and associated, such data may provide inconsistent knowledge. Ensure that we derive only consistent and useful knowledge is a major challenge in this context.

Deriving new knowledge is always a challenge, but discovering relevant, accurate, consistent and compact knowledge (of real use) can be daunting in a big and heterogeneous domain.

Furthermore, explaining abnormal events in a big and heterogeneous data context requires appropriate relevant, accurate, consistent and compact knowledge.

This approach has been applied successfully in cities. It is extensible and scalable for reasoning over heterogeneous and big data from any domains, including transportation, energy and many more.


Freddy Lecue

Example of results

We conducted three pilot projects in the transportation domain. Shown here are selected results for Dublin.

Piloted in three cities - Dublin