Bootstrapping knowledge representations - from entailment meshes via semantic nets to learning webs

摘要

The symbol-based epistemology used in artificial intelligence is contrasted with the constructivist, coherence epistemology promoted by cybernetics. The latter leads to bootstrapping knowledge representations, in which different parts of the system mutually support each other. Gordon Pask's entailment meshes are reviewed as a basic application of this approach, and then extended to entailment nets: directed graphs governed by the "bootstrapping axiom", determining which concepts are to be distinguished or merged. This allows a constant restructuring of the conceptual network. Semantic networks and frame-like representations can be expressed in this scheme by introducing a basic ontology of node and link types. Entailment nets are then generalized to associative networks with weighted links. Learning algorithms are presented which can adapt the link strengths, based on the frequency with which links are selected by hypertext users. It is argued that such bootstrapping methods can be applied to make the World Wide Web more intelligent, allowing it to self-organize and support inferences.