LINGUIST List 12.1241

Fri May 4 2001

Confs: Open-Domain Question Answering/ ACL-2001

Editor for this issue: Lydia Grebenyova <lydialinguistlist.org>

Directory

1. Priscilla Rasmussen, ACL-2001 Workshop on Open-Domain Question Answering

ACL-2001 Workshop on Open-Domain Question Answering - schedule

9:00 - 10:00 Welcome + overview
10:00 - 10:25 	E. Breck. M. Light, G. Mann. E. Riloff, B. Brown, P.
			Anand, M. Rooth and M. Thelen
 	"Looking under the Hood: Tools for Diagnosing your
			Question Answering Engine"
10:25 - 10:50 	H. Kim, K. Kim, G.G. Lee and J. Seo
 	"MAYA: A Fast Question-Answering System Based on a
			Predictive Answer Indexer"
10:50 - 11:15 	Break
11:15 - 11:40 	U. Hermjakob
 	"Parsing and Question Classification for Question
			Answering"
11:40 - 12:05 	G. S. Mann
 	"A statistical method for short answer extraction"
12:05 - 12:30 	R. Zajac
 	"Towards Ontological Question Answering"
12:30 - 1:45 	Lunch
1:45 - 2:10 	M.Pasca and S.Harabagiu
 	"Answer Mining from On-Line Documents"
2:10 - 2:35 	O. Ferret, B. Grau, M. Hurault-Plantet, G. Illouz and
			C. Jacquemin
 	"Terminological variants for document selection and
			question/answer matching"
2:35 - 3:30 	A. Fujii and T. Ishikawa
 	"Question answering using encyclopedic knowledge
			generated from the Web"
3:30 - 4:00 	Break
4:00 - 5:00 Keynote Speech, John Prange, ADRA/AQUAINT program
 		"Advanced Question and Answering:The Right Problem at 
			the Right Time or a Bridge Too Far?" 
		 (see abstract below)
5:00 - 5:45 Panel, John Prange, Donna Harmann, Sanda
			Harabagiu, Yael Ravin
 	"The Future of QA"


"Advanced Question and Answering: The Right Problem at the Right Time or a
Bridge Too Far?"

 Dr. John D. Prange
 Technical Director
 Advanced Research and Development Activity in Information Technology

 Fort George G. Meade, Maryland, USA

 ABSTRACT

 Over the past 50 years various fields within human language
technologies and the information sciences have occasionally been presented
with grand challenge problems that have generated enormous interest and
enthusiasm. Almost without exception, researchers and technologists have
embraced these challenges and have expended significant intellectual
capital against the obstacles, quagmires, and barriers that these challenge
problems have presented. If these obstacles, quagmires, and barriers were
substantially overcome, a major breakthrough or advance typically occurred
in the field. In turn these successes produced substantial new knowledge,
insights, and understanding, and they can open the doors to previously
unthought of opportunities and applications. But if the challenge proved
to be too formidable then the field often suffered a major setback and an
intellectual "winter" frequently ensued. In either case the field was
forever changed. In essence the challenge problem becomes a watershed
event.

 We believe that "Advanced Question and Answering" has the potential of
being a watershed, Grand Challenge Problem for multiple fields; to include
information retrieval, natural and statistically-based language processing,
computational linguistics, knowledge acquisition and representation,
machine learning, dialog, summarization, and natural language generation.
In this talk we will describe our interpretation of "Advanced Question and
Answering" as a Grand Challenge Problem. We will identify what we believe
are the key elements and components of Advanced Question and Answering and
what obstacles, quagmires, and barriers must be overcome if we are to
collectively achieve substantial success against this Grand Challenge
Problem. In particular we will specifically highlight and discuss the
following:

 Multi-dimensionality of both Questions and Answers
 Role of Knowledge
 Role of Context
 Modeling/Profiling the User
 Overcoming the Data Chasm, to include:
 Highly heterogeneous data and information sources
 Data Volumes
 Missing Data
 Conflicting Data
 Understanding and Fusing Information across "documents"
 Simultaneously achieving Accuracy, Completeness, Usability, and
 Timeliness
 Role of component and system-level evaluation

 We will conclude by offering our answer to the question posed in the
title of this talk: "Advanced Question and Answering: The Right Problem at
the Right Time or a Bridge Too Far