Research in Knowledge Representation for Natural Language Understanding
Abstract:
BBNs ARPA project in Knowledge Representation for Natural Language Understanding is aimed at developing techniques for rendering computer-based assistance to a decision maker who is attempting to understand and react to a complex, evolving situation. This report summarizes our research in knowledge representation an natural language over the length of the contract. In particular, we discuss the development if KL-ONE, NIKL, and KL-TWO, the advances in RUS and the PSI-KLONE systems, and our work in understanding discourse and speakers intentions. Key words include Artificial intelligence, natural language understanding, knowledge representation, semantics, semantic networks, KL-TWO, NIKL, belief and knowledge, reasoning, RUP, syntax, parsing, reference, miscommunication, plan recognition, speaker intention, planning, situation calculus, discourse, semantic acquisition, HSP, IRACQ.