Information Consolidation in Large Bodies of Information
Gerhard Wurzinger (Graz University of Technology, Austria)
Abstract: Due to information technologies the problem we are facing today is not a lack of information but too much information. This phenomenon becomes very clear when we consider two figures that are often quoted: Knowledge is doubling in many fields (biology, medicine, computer science, ...) within some 6 years; yet information is doubling every 8 months! This implies that the same piece of information/knowledge is published a large number of times with small variations.
Just look at an arbitrary news item. If considered of some general interest reports of it will appear in all major newspapers, journals, electronic media, etc. This is also the problem with information portals that tie together a number of large databases.
It is our contention that we need methods to reduce the huge set of information concerning a particular topic to a number of pieces of information (let us call each such piece an "essay" in what follows) that present a good cross-section of potential points of view. We will explain why one essay is usually not enough, yet the problem of reducing a huge amount of contributions to a digestible number of essays is formidable, indeed is science fiction at the moment. We will argue in this paper that it is one of the important tasks of computer sciences to start tackling this problem, and we will show that in some special cases partial solutions are possible.
Keywords: information consolidation
Categories: H.3.4, H.3.5, H.3.7, H.4.3, H.5.1, M.1