Visualization and Manipulation of Incomplete and Uncertain Dependencies by Decision Diagrams

Denis V. Popel (Department of Computer Science, Baker University, USA)

Abstract: The data mining community is focused on a variety of methods and algorithms to manipulate incompletely specified or uncertain data and their dependencies. The major obstacle in the representation and visualization of incompletely specified data is the size explosion problem through defining undefined or uncertain values, which commonly raises questions about suggested heuristics and their practical applicability. Recently, there is a renewed interest in resolving the size explosion problem for incompletely specified and uncertain data based on symbolic techniques. One of such techniques, decision diagram, has been successfully applied to many knowledge visualization and data manipulation problems.

Keywords: data mining, decision diagrams, incompletely specified functions, minimization

Categories: F.4.3, I.6.8