Since the recent past, the data warehousing and BI technologies are being extended towards the analysis of new types of data, including among others, images, spatial, texts, XML, graphs, and sequences. Data of these types are ubiquitous nowadays and their amount grows fast. The analysis of network link traffic, sensor data, or large social and information networks (e.g., Facebook, Twitter, etc.), credit card fraud detection, share price changes, and real-time Web analytics (e.g., for ad campaign effectiveness or click-streams) are prominent examples of the application of modern BI systems.

Recently, the analysis of sequential data turned out to be one of the hot research topics. The techniques allowing to analyse sequential data are very important in public transportation, sensor installations, and click-stream analysis. The existing commercial BI products do not support support the analysis of sequential data. So far, a few research contributions have been proposed and they offer only the basic functionalities. For this reason, there is an evident need to conduct research in the area of sequential BI towards the development of the five following components: (1) a fully functional query language, (2) a user interface for querying and visualising sequential data, (3) a query optimiser and query optimisation techniques, (4) data structures for storing and querying efficiently sequential data, (5) techniques for mining complex patterns in sequential data.

The aim of this topic is to develop a comprehensive solution for sequential BI, including the aforementioned four components of sequential BI.

Main Advisor at Aalborg Universitet (AAU)
Co-advisor at Poznan University of Technology (PUT)