DB group
 

Research Profile

The main research field of the group is the management and retrieval of data and information. Our research activities are based on the two complementary technologies of database systems where the primary focus is on the management, storage and querying of structured data, and information retrieval that focuses on indexing and retrieval of unstructured information in the context of user needs.

Our research approaches are within the areas of problem solving methodology, with a focus on analysis and design of innovative solutions, development of experimental prototype algorithms and systems, testing and evaluation.

Research Activities

Currently the group has research activities in (but not limited to) the following areas:

  • Distributed and parallel database systems with a focus on performance, scalability, availability and collaboration. Focus in recent years has been on efficient algorithms for distributed query processing.
  • Query processing with a focus on next generation search engines, rank-aware query processing, fusion of text search and spatial query processing and applications of database techniques to similarity search.
  • Information retrieval with a focus on application/integration of IR-techniques in databases and digital libraries, including developing efficient index structures and optimized query processing, and IR-theories, aiming at developing novel, effective and efficient retrieval approaches and methods for all types of information and search strategies.
  • Text mining with a focus on extraction of concepts and entities in unstructured text to improve retrieval and discovery of relevant information. As an example, our activities in this area include temporal analysis of text documents, spatio-temporal analysis of image annotations, and using named-entity recognition in document ranking and retrieval.
  • Digital Libraries with a focus on methods and tools to make digital, cultural and scientific content available, searchable and accessible in the context of user needs and for reuse in a broader context in the semantic web of the future.

As part of our activities in the above areas, members of the group are engaged in a range of research projects and publish in relevant venues such as renowned conferences and journals. We are also responsible for giving courses in topics related to our research, ranging from basic undergraduate courses to advanced research-based post-graduate courses. And, throughout the years, our group has graduated a large number of master and doctoral candidates.

During the last decade there has been a trend toward convergence of database systems and information retrieval, e.g. in the way information retrieval and search technologies have revitalized research in database querying. Considering our competence we particularly aim to contribute to the overlapping areas. One example of a project where we have been able to realize a synergistic potential is the information Access Disruptions project (iAD) where we have utilized techniques from (database) query optimizing in the realm of similarity search.

The group has currently research collaboration (in projects, publications, extended visits) with a large number of universities and research institutions, the most important in terms of cooperation being the Cornell, University of California Berkeley, University of California Irvine, and Athens University of Economics and Business.

Our group has a long tradition in working with industry-relevant problems, which has led to several spin-off companies, and we intend to continue this tradition. Currently, we have collaboration with the Trondheim department of Oracle (which is a result of a spin-off company from the distributed and high availability activities in our group). We have also collaboration with Microsoft Enterprise Search (formerly FAST Search and Transfer). The Data and Information Management group has also an extensive collaboration with other partners in financed research projects.