Weakly-supervised machine-learning techniques for Appraisal analysis

---------------------------------------------------------------------------
                ***Joint Seminar***
---------------------------------------------------------------------------
The Hong Kong University of Science & Technology

Department of Computer Science and Engineering
Department of Electronic and Computer Engineering
Human Language Technology Center
---------------------------------------------------------------------------

Speaker:	Jonathon READ
		Department of Informatics
		University of Sussex

Title:		"Weakly-supervised machine-learning techniques
		for Appraisal analysis"

Date:		Monday, 1 December 2008

Time:		4:00pm - 5:00pm

Venue:		Lecture Theatre F
		(Leung Yat Sing Lecture Theatre, near lifts25/26),
		HKUST

Abstract:

The Appraisal framework is a theory of the language of evaluation,
developed within the tradition of Systemic Functional Linguistics. The
framework describes a taxonomy of the types of language used to convey
evaluation and position oneself with respect to the evaluations of other
people. Accurate automatic recognition of these types of language can
inform an analysis of document sentiment.

I will describe the preparation of a corpus of book reviews manually
annotated with labels from the Appraisal framework.  The difficulty of the
task is assessed by way of an inter-annotator agreement study.  This
corpus forms test data for weakly-supervised algorithms employing measures
of semantic similarity that automatically label words according to their
Appraisal type.  I will present an evaluation of these methods, discuss
potential improvements and outline possible applications.


**********************
Biography:

Jonathon READ is currently concluding a doctoral programme of research at
the University of Sussex under the supervision of Professor John Carroll.
During this programme he has also taken a place on the student board of
the European Association for Computational Linguistics, and carried out
consultancy work in the areas of sentiment analysis, social network
analysis and topic classification.  For further details please see
http://www.informatics.sussex.ac.uk/users/jlr24/.