In last few years it became clear that the big amount of data, produced by all software used every day
by all kind of organizations, need to be managed and analyzed correctly and taking advantage of all
their particular characteristics. In such a way, data become a new asset useful for supporting
organizations work and decision makers. For this purpose, different technologies and concepts may be
used, including data warehouses, business intelligence and statistical analysis. Despite these
technologies are frequently used by same people with a same goal, they are independent and with
disjoint sets of capabilities, each one suffering the lack of features of the other. For example, OLAP
analysis allows one to navigate through and aggregate data on the basis of different dimensions, but it
lacks of advanced statistical capabilities as well as the possibility to represent and manage temporal
data and dimensions in order to better discover (temporal) trends and patterns. On the other hand,
statistical analysis provide a lot of tools for discovering trends and patterns, for preprocessing and
getting derived data, and for aggregate data, but it is not able to produce rich interactive reports as
OLAP tools do.
This project aims to develop and design a set of concepts and tools for (1) modeling multi-granular
temporal data in data warehouses and (2) integrating OLAP techniques and statistical analysis in order
to better exploit the analysis capabilities of both technologies and provide an unified tool.