CLARET: A new Relational Learning Algorithm for Interpretation in Spatial Domains

A new learning algorithm: Consolidated Learning Algorithm based on Relational Evidence Theory (CLARET) is presented, which integrates a machine learning methods with graph matching techniques. A systematic approach to nding interpretations is demonstrated for an on-line, hand drawn, schematic diagram and symbols recognition system. The approach uses an adaptive representational bias and search strategy during learning by e ciently grounding the learning procedures in the relational spatial constraints of their application.