VLDB Journal Special Issue on


            Integration of Databases and Information Retrieval


                          Call for Contributions


 The fields of Databases (DB) and Information Retrieval (IR) have a long
tradition going back to the sixties. At this time the distinction was
clear: Database systems were for records and information retrieval systems
were for collections of textual documents. These early origins have led to
the typical view that databases manage ~structured~ data while information
retrieval supports ~unstructured data~, and to the separation into two
fields with largely independent research agendas. There were attempts to
bridge the gap between the fields and to integrate DB&IR in the late
seventies and early eighties, but these ran into a number of problems, both
from the theoretical and implementation perspectives. Even at that time,
however, it was clear that most information had both structured and
unstructured components and neither could be ignored. In recent years, very
much as a consequence of both the Web, with huge amounts of primarily
textual information, and of XML with its flexible data model, the
separation of databases and information retrieval has become obsolete. Many
researchers consider the integration of DB&IR an extremely important topic
for the future of information systems, and evaluation forums like INEX are
providing new venues to study this issue.


In order to reflect this new awareness, the VLDB Journal will publish a
special issue devoted to the IR & DB integration. Therefore we ask for
submissions to be considered for publications in this special issue that
will bring together new research results that are not yet published
elsewhere and that will stimulate further research.


TOPICS OF INTEREST


Relevant topics include, but are not limited, to the following:
      Architecture: preprocessor solutions, middleware, or systems from
      scratch?
      DB&IR query languages and APIs (browsing, navigation, exploration,
      annotation, further result processing)
      Retrieval models and ranking functions
      DB&IR query paradigms, query generation, expansion, relevance
      feedback
      Query strategies and optimization over search engines
      Feature extraction in DB&IR, assignment of weights or importance to
      data, learning from large databases
      Evaluation of DB&IR systems
      Information visualization, abstracting, aggregation
      Applications of DB&IR


GUEST EDITORS


Hans-Jörg Schek, ETH Zurich, Switzerland and UMIT Hall, Austria
(schek@inf.ethz.ch )


Bruce Croft, University of Massachusetts, Amherst, USA (croft@cs.umass.edu)


IMPORTANT DATES:


  * Paper submissions: 15 September, 2006


 * First round notification: 15 December, 2006


 * Revised: 15 February, 2007


 * Second round notifications: 15 March, 2007


 * Final version: 30 March, 2007