Data Mining and Knowledge Discovery by Means of Monotone Boolean Functions
In all previous discussions the problem was how to infer a general Boolean function based on some training examples. Such a Boolean function can be completely inferred if all possible binary examples (states) in the space of the attributes are used for training. Thus, one may never be 100% certain about the validity of the inferred knowledge when the number of training examples is less than 2 n . The situation is different, however, if one deals with the inference of systems that exhibit monotonic behavior. The developments presented in this chapter are based on the award-winning doctoral work of Vetle I. Torvik and in particular on the research results first published in [ Torvik and Triantaphyllou, 2002; 2003; 2004; 2006].
KeywordsBoolean Function Query Complexity Conjunctive Normal Form Inference Algorithm Inference Process
Unable to display preview. Download preview PDF.
- Judson, D.H., (2001), “A Partial Order Approach to Record Linkage,” Federal Committee on Statistical Methodology Conference, November 14–16, Arlington, VA, U.S.A.Google Scholar