Skip to main content

CIC Seminar March I

Date: March 15, 2017

Time: 1.30 – 2.30pm, followed by afternoon tea

Location: CLT Learning Space, Building 105, Room 107

Speaker: Professor John Bateman

Title: Learning to count: the challenge of turning rich multi semiotic materials into (audio-) visual data (and then turning it back again…)

RSVP: Please RSVP via Eventbrite on this link by COB Monday 13 March, 2017

Abstract:

One of the primary challenges with dealing with the rich multimodal content taken for granted in many of today’s communicative practices is being able to focus in on what is meaningful. Audio-visual data generally contains such a wealth of distinctions that is by no mean straightforward to know which dimensions of variation are going to be useful for interpretation and which not. In this talk, I discuss how more detailed and theoretically motivated accounts of the multimodal meaning-making process can help with this task. Particular communicative artefacts draw on specific meaning-making practices that allow material distinctions to be assigned meanings. Building empirically-motivated top-down constraints of this kind into automatic interpretation processes may open the way to considerable improvements in our ability to process rich sources of data.

Bio:

Professor John Bateman PhD is Professor of Applied Linguistics in the English and Linguistics Departments of the University of Bremen, specialising in functional, computational and multimodal linguistics. He has been investigating the relation between language and social context for many years, focusing particularly on accounts of register, genre, functional variation, multilingual and multimodal linguistic description, and computational instantiations of linguistic theory. He has published widely in all these areas, as well as authoring several introductory and survey articles on natural language generation, systemic-functional linguistics and multimodal analysis. His current interests centre on the application of functional linguistic and corpus methods to multimodal meaning-making, analysing and critiquing multimodal documents of all kinds, the development of linguistically-motivated ontologies, and the construction of computational dialogue systems for robot-human communication.