In this paper, we present an inference mechanism called the integrated parsing engine which provides a uniform abductive inference mechanism for natural language understanding. It can (1) make plausible assumptions, (2) reason with multiple alternatives, (3) switch the search process to the maximally plausible alternative, (4) detect contradiction and tame con-clutions which depend on inconsistent assumptions, and (5) update plausibility factor of each belief based on new observations. We demonstrate that a natural language understanding system using the integrated parsing engine as a subsystem can pursue a guided search for most plausible interpretation by making use of syntax, semantics, and contextual information.
展开▼