Are frames adequate for capturing language meaning? A discussion of chained metonymy and spreading activation.

Add to your list(s) Download to your calendar using vCal

• Dr. Michael Moodie
• Friday 14 March 2014, 16:00-17:00
• G26, Mechanical Engineering.

If you have a question about this talk, please contact Leandro Minku.

Host: John Barnden

If frame semantics is not up to the job, what is ? I propose to discuss the merits of clustering at some length. Sentences which do not permit inference, paraphrase or near-synonymy are probably inadequate when we look at whole texts. Examples will be provided and elicited from the attendees (I speak as a Gricean) to provide wide coverage from the lexicon. This can include topics like: CLAY and PRESSING , KNIVES and CUTTING , CORN and GRINDING . Provisionally, I suggest discussing these stripped of modals, ambiguity, lying, etc. No area is out of bounds. Something to consider is how our communication would be impeded if we were prohibited from making any reference to POSSESSION (including the verb HAVE ) or to the cluster round SAFETY -HARM. Reference will be made to Fillmore’s Verbs of Judging (1971) and Case for Case (1968). The work of van Dijk and Kintsch in the 70s will be mentioned. Any queries arising out of FrameNet work will be welcome. Possibly, there will be time to look at the significance of different paraphrases, at the challenges in learning Spanish from Mrs Caceres (who spoke no English), at child language, at a little story about the social significance of ALCOHOLISM , and at the meaning and translation of “Claro que si!”.

This talk is part of the Artificial Intelligence and Natural Computation seminars series.

This talk is included in these lists:

• Artificial Intelligence and Natural Computation seminars
• Computer Science Departmental Series
• Computer Science Distinguished Seminars
• G26, Mechanical Engineering
• computer sience

Note that ex-directory lists are not shown.