Effective supra-classifiers for knowledge base construction

We explore the use of the supra-classifier framework in the construction of a classifier knowledge base. Previously, we introduced this framework with in which labels produced by old classifiers are used to improve the general ization performance of a new classifier for a different but related classif ication task (Bollacker and Ghosh, 1998). We showed empirically that a simp le Hamming nearest neighbor is superior to other techniques (e.g., multilay er perception (MLP), decision trees, Naive Bayes, Combiners) as a supra-cla ssifier, Here, we describe theoretically how the probability that the Hammi ng nearest neighbor supra-classifier will predict the true target class app roaches certainty at an exponential rate as more classifiers are reused. Th e scalability of the Hamming nearest neighbor with large numbers of previou sly created classifiers makes it a good choice as a supra-classifier in the application of building a repository of domain knowledge organized as a cl assifier knowledge base. (C) 1999 Elsevier Science B.V. All rights reserved .