Document Type

Student Presentation

Presentation Date

4-17-2017

Faculty Sponsor

Casey Kennington

Abstract

Spoken conversation between two present interlocutors is the foundation of natural language interaction and the source for child language acquisition (Fillmore, 1981; McCune, 2008). This research aims to model this natural system using the words-as-classifiers (WAC) model of grounded lexical semantics (Kennington and Schlangen, 2015), to test if a replicated environment can produce similar acquisition results in a machine robot (Anki’s Cozmo). The main question is whether meaning can be acquired and represented using the semantic grounding provided through the WAC model. The results of this research will expand our understanding of lexical meaning and language interaction, which has implications in all areas of data science and natural language processing.

References:

Charles J. Fillmore. Pragmatics and the description of discourse. Radical pragmatics, pages 143–166, 1981.

Casey Kennington and David Schlangen. Simple Learning and Compositional Application of Perceptually Grounded Word Meanings for Incremental Reference Resolution. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 292–301, Beijing, China, 2015. Association for Computational Linguistics.

Lorraine McCune. How Children Learn to Learn Language. 2008.

Share

COinS