A Quasi-Metric for Machine Learning

摘要

The subsumption relation is crucial in the Machine Learning systems based on a clausal representation. In this paper we present a class of operators for Machine Learning based on clauses which is a characterization of the subsumption relation in the following sense: The clause C_1 subsumes the clause C_2 iff C_1 can be reached from C_2 by applying these operators. In the second part of the paper we give a for-malization of the closeness among clauses based on these operators and an algorithm to compute it as well as a bound for a quick estimation.