A learning mechanism for the selection of hypotheses on abductive reasoning

We propose a learning mechanism to learn how to select hypotheses from a set of abducibles (possible hypotheses) on abductive reasoning. Abductive reasoning is to infer an explanation of why observations could abduction this explanation is selected from a set of the given possible hypotheses. This selection follows the plausible heuristics (ME (Minimal Explanation) criterion, LPE (Least Presumptive Explanation) criterion, or Basic criterion). Abduction is characterized by these semantical choosing principles which is different from the MDL on induction. This learning mechanism is to learn preferentially propositions or rules that are selected by the heuristics. We try to integrate abductive learning and inductive learning by the number of examples for learning.