Toward a Categorical Link Grammar for Knowledge Piece Generation in Emergency Decision-Making Support

Abstract

Emergency knowledge pieces are special knowledge units with categorical structures that keep comparatively sufficient semantic information for advanced knowledge reorganization; however, the primary question is how to generate knowledge pieces from textual data for more strict knowledge processing. Link Grammar is a typical method on parsing and its special link mechanism of using connectors give us an inspiration to extend this method to the knowledge piece generation. Drawing on the link mechanism, we propose a Categorical Link Grammar (CLG) that extend the connectors and their formulas to the categorical types and their order formulas, and give the flow of generating knowledge pieces.