论文信息 - A Cognitive-Based Approach to Learning Integrated Language Components

A Cognitive-Based Approach to Learning Integrated Language Components

The learning component of a cognitive-based language model (LEAP) designed to easily integrate into agent systems is presented. Building on the Interlaced Micro-Patterns (IMP) theory and the Alchemy/Goal Mind environment, the LEAP research improves agent-to-human and agent-to-agent communication by incorporating aspects of human language development within the framework of general cognition. Using a corpus of child through youth fiction, we provide evidence that micro-patterns can be used to simultaneously learn a lexicon, syntax, thematic roles and concepts.

Charles Hannon | Jonathan Clark

[1] A. Newell. Unified Theories of Cognition , 1990 .

[2] C. Martindale,et al. Cognitive Psychology: A Neural-Network Approach , 1990 .

[3] Diane J. Cook,et al. Exploring the Use of Cognitive Models in AI Applications Using the Stroop Effect , 2001, FLAIRS.

[4] Peter Wagner,et al. An Interactive Dialogue System for Knowledge Acquisition in Cyc , 2003, IJCAI 2003.

[5] C. Hannon,et al. Emotion-based Control Mechanisms for Agent Systems , 2003 .

[6] Diane J. Cook,et al. Developing a Tool for Unified Cognitive Modeling Using a Model of Learning and Understanding in Young Children , 2001, Int. J. Artif. Intell. Tools.

[7] Michael D. Byrne,et al. ACT-R/PM and menu selection: applying a cognitive architecture to HCI , 2001, Int. J. Hum. Comput. Stud..

[8] Charles Hannon,et al. A Geographically Distributed Processing Environment for Intelligent Systems , 2002 .

[9] C. Lebiere,et al. The Atomic Components of Thought , 1998 .

[10] Christiane Fellbaum,et al. Book Reviews: WordNet: An Electronic Lexical Database , 1999, CL.

[11] John R. Anderson. Cognitive Psychology and Its Implications , 1980 .