Upload
keaton
View
24
Download
0
Tags:
Embed Size (px)
DESCRIPTION
German Competence Center in Speech and Language Technology. J. Capstick, T. Declerck, G. Erbach, A. Jameson, B. Jörg, R. Karger, H. Uszkoreit, W. Wahlster, T. Wegst. LREC, Las Palmas, 30 May 2002. Project COLLATE. - PowerPoint PPT Presentation
Citation preview
German Competence Center
in Speech and Language Technology
J. Capstick, T. Declerck, G. Erbach,A. Jameson, B. Jörg, R. Karger,H. Uszkoreit, W. Wahlster, T. Wegst
LREC, Las Palmas, 30 May 2002
Project COLLATE
Theme: Computational Linguistics and Language Technology for
Real World Applications
Support: A Grant by the
German Federal Ministry for Education and Research (BMBF) for RTD
strengthening the position of Saarbrücken as a Competence
Center for Language
Technology
PIs: Hans Uszkoreit, Manfred Pinkal and Wolfgang Wahlster
Duration: Spring 2001 - end of 2003
COLLATE Project Structure
informationextractionand fusion
dialogue forknowledge
access
informationmanagementand retrieval
basic functionalities of LT-basedknowledge management
Virtual Information
Center
Demonstration Center
EvaluationCenter
Competence Center in Speech and Language Technologze
Coordination, Controlling, Reporting, WWW Presence, PR, EventsCoordination, Controlling, Reporting, WWW Presence, PR, Events
Need for a Competence Center
LT field is growing Researchers from LT and neighbouring disciplines need a
comprehensive information service Users of LT need in-depth information about technologies,
available products and suppliers Users of LT need support to find or develop solutions
meeting their application requirements Developers and users of LT need criteria and evaluations
regarding the usability of LT in real world applications Students of LT need an information source
1. Virtual Information Center: LT World
LT World: Idea and Context
The virtual information center is a comprehensive WWW-based information and knowledge service for the entire area of language technology.
LT World is a “virtual” center in the sense that most information will physically remain with their creators or with other service providers.
The virtual information center has been online since October 2001 under the name „LT World“ for „Language Technology World“ (www.lt-world.org)
Virtual Information Center - LT World
Information and Knowledge
Technical and Scientific Information
Players and Teams
Persons, Projects, Organisations
Resources and Results
Research Systems, Commercial Products
Communication and Events
News, Conferences
LT World Ontology
Publications
Products Projects People
Layer 2: Specific Ontologies
Corpora etc.
Layer 1: Dublin Core
Layer 3: Ontology for CL & LT
LT World: Coverage
99 topic nodes
300 NLP tools and products
1800 people
850 organisations
500 projects
Data Acquisition Process
Manual collection, categorization and annotation of URLs by students and staff
Sources: conference proceedings and journals, lists of links on the web,
Self-registration and correction of data by users of the service
Technical/scientific information in topic nodes has been provided by domain experts
LT World: Topic Nodes
Topic nodes are the main information unit of the Area “Knowledge and Information”. They are organized in a shallow slightly multidimensional hierarchy following the chapter plan of the second edition of the Language Technology Survey
Example of the shallow hierarchy
Information Extraction• Named Entity Recognition
• Terminology Extraction
• Relation Extraction
• Answer Extraction
Information for each Topic
Name
Acronyms
aka‘s, Term Translations
Short Definition
Overview Article (from HLT Survey)
Topic Websites
R&D Prototypes/Products
Projects
People
Literature
Hyperlinking between Sections
Relationship to External Resources
Included but autonomous resources: ACL Software Registry, Language Technology Survey
Systematically cross-Linked and Cross-Searchable Resources: all OLAC Resources such as (LDC, ELRA , SIL, ACL SR, and OLAC Home)
Systematically crosslinked resources: HLT Central, ELSNET, ACL NLP Universe, EACL, COLIBRI
Linked resources: All other relevant resources relevant for LT
Future Work: Virtual Information Center
Update of Information and Knowledge section in cooperation with 2nd edtion of HLT Survey
Interaction (chatrooms, discussion boards) Job offers
Use of language technologies to improve the content of LT World Improved hyperlinking between the different sections Resource discovery Automatic metadata extraction Construct corpus of LT area as R&D resource
2. Demonstration Center
Demonstration Center
Ppotential users and other interested parties can see and test the most important research prototypes and products of language technology
The demo center is available for seminars, tutorials, and information visits
Beneficiaries are: companies and other organizations interested in the deployment of language technology, researchers and developers of language technology, other decision makers with an interest in the state of the art in LT
Demonstration Center: Technical Setup
PC-based demonstration kiosks
Audio-visual network allows redirection of voice and video in/output between kiosks
Specially equipped room for in-depth demonstrations
Demo scripts and data for different kinds of applications
Demonstration Center: Installed Software
Machine Translation (LOGOS, Linguatec Personal Translator ...)
Spoken dialogue system (Sympalog, Nuance, Speechworks...)
Text to speech synthesis (RealSpeak, Mary ...) Information Extraction (LaSIE, SPPC, ...) Dictation System (Dragon NR, L&H ASR 16, ...) Finite State Tools (XEROX XLE ...) Multimedia Indexing and Search (MUMUS ...) Spellchecking for professional users (CLT Corrigo ...) Voice Dialling (VoiceDirector)
.....
Future Work: Demonstration Center
New software is acquired and existing software updated
Information days and seminars (starting autumn 2002)
3. Evaluation Center
Evaluation Center: Idea
Development of methodology for an individual customized evaluation of LT applications
The focus is neither general technology evaluation (TREC, TIPSTER etc.) nor generic product testing according to a fixed set of criteria
The focus of the evaluation center is a thorough, customized and goal specific evaluation of individual applications. The evaluation will center on usability, interoperability, and adequacy with respect to specified tasks.
Evaluation Center: Usability Studies
Evaluation of the overall usability of LT systems for real users in typical contexts
Current focus is on studying the interaction of users with LT applications, making use of a portable eye-tracker
Combination of eye-tracking and user interviews yields new insights into usability design
Initial Version of User Interface
Improved Version of User Interface
Conclusion
Thanks to the funding by BMBF and thanks to existing valuable resources such as the HLT Survey and the ACL Software Registry, we have set up the most comprehensive information resource on Language Technology.
We welcome any comments and proposals for collaboration that could help to strengthen LT World and the entire infostructure of our field.
Demonstration center and evaluation center provide useful services to developers and users of LT.
http://www.lt-cc.org/