|
|
Home
Page |
Dr Mick O'Donnell,
Computational Linguist
|
|
Research Interests
|
Text Generation (Composing text from
underlying databases). |
|
Sentence Generation (Generation of
individual sentences). |
|
Sentence Analysis (Analysis of individual
sentences using sSystemic
formalism). |
|
Knowledge Representation (representation
of conceptual and linguistic
data). |
|
Linguistic Coding Tools: Tools for
marking up text structure
or features (see below). |
|
Interaction Structure: modelling the
structure of human interaction,
including exchange (conversation) structure, task (generic) structure.
Including
both verbal and nonverbal behaviour. Dynamic modelling of interaction. |
|
Systemic Formalism: using the Systemic
formalism to model all
information (language, behaviour, knowledge, etc.). Notion of strata,
inter-stratal
mapping. Systemics as a knowledge-representation language. |
Projects
|
ILEX
(Intelligent Label
Explorer) (with Chris Mellish, Jon Oberlander and Alistair Knott) seeks
to automatically generate labels for items in an electronic catalogue
(or museum gallery) in such a way as to reflect the interest of the
user and also opportunistically
to further certain educational (or other) aims. This project is in
collaboration
with the National Museums of Scotland, Interactive Information and VIS
Interactive
Media. For more information on this project (papers, etc.), and to
access
our prototype system, click here. |
|
HIPS
(Hypernavigation
in Physical Space) (European Consortium, with Chris Mellish, Jon
Oberlander
and Marc Moens) Developing handheld devices to deliver dynamically
adapted
presentations as the visitor walks through an exhibition space. For
more
information on this project (papers, etc.), click here. |
|
M-Piro
(Multilingual Personalised
Information Objects) (European Consortium) Project to deliver a
multilingual
version of ILEX-type system. I worked on the project in the first year,
to
extend ILEX's multilingual capabilities. |
|
AcaMed I was
contracted by Language & Computing
nv. to manage a
team of researchers developing natural language understanding
capabilities
for texts in the medical domain. This is mainly funded under a Flemish
Government
Research Grant for the ACAMED-project. ACAMED is a generic tool for
automatic
ICD-code extraction from medical free texts. |
Information
Personal
Contact
Email: micko@wagsoft.com.
Phone Wk/Hm: 0034-918031281
Mail: Calle Aurora, 3, 5A,
28760, Tres Cantos (Madrid),
Spain.
|
|