MS.14. Active Business Intelligence

Whereas traditional Business Intelligence solutions focus on the setting where one has to analyze historical data in order to evaluate potential strategic business options, there is a growing need for business intelligence solutions that are (re)active, i.e., can deal with continuously updated data coming from distributed sources at unpredictable rates. (Examples include financial stock analysis, traffic and mobility management, click-stream inspection, ...).

These requirements have led to the development of systems that are specifically designed to process information as a flow (or a set of flows) of data items according to a set of pre-deployed processing rules. However, as Cugola and Margara point out in their 2012 Computing Survey article, "despite having a common goal, these systems differ in a wide range of aspects, including architecture, data models, rule languages, and processing mechanisms. In part, this is due to the fact that they were the result of the research efforts of different communities, each one bringing its own view of the problem and its background to the definition of a solution."

As a result there is currently a lack of a unified model underlying these systems, both in terms of:

  • rule language features (from languages that have many aggregation operators but limited operators for sequencing to langauges that are based on automaton models with advance sequencing and filtering but with limited aggregation),
  • processing mechanisms (solutions typically focus on the non-distributed or non-parallel environments), and
  • optimization strategies.

In contrast, if active business intelligence is to become mainstream, such a unified model is mandatory.

The objective of this research project is therefore to (1) design a rule language that encompasses and combines existing solutions in terms of expressiveness; (2) provide a general, unified execution model for this language, targetting distributed and multi-core clusters; (3) designing corresponding unified scalable execution and rule optimization strategies.

Main Advisor at Université Libre de Bruxelles (ULB)
Co‐advisor at Aalborg Universitet (AAU)