Member Login

JCal Pro Mini-calendar

October 2019 November 2019 December 2019
Mo Tu We Th Fr Sa Su
Week 44 1 2 3
Week 45 4 5 6 7 8 9 10
Week 46 11 12 13 14 15 16 17
Week 47 18 19 20 21 22 23 24
Week 48 25 26 27 28 29 30

Current Time

The time is 20:05:41.
Niels Beuck Print E-mail

 

Niels Beuck


Phone:  +49 40 42883-2363
Room:  F-409, Informatikum
eMail: beuck \at informatik.uni-hamburg.de
   
Supervisor: Prof.  Dr. Wolfgang Menzel
   

 

Short Biography

Niels Beuck is a PhD Candidate in the Natural Language System Group NATS at the Department of Informatics of Hamburg University. He joined CINACS in July 2009.

Before that he studied Informatics in Hamburg, where his research focused on incremental parsing of natural language. 

 

Publications

publication site

 

Research Topic

Anticipatory Incremental Language Processing in Multi-Modal Context

The starting point of my project is the observation that human language processing is
inherently incremental, i.e. a sentence is not processed as a whole but each part of it is analysed once it is available. This processing mode allows humans to use the time an utterance takes to be produced, to already analyse and react to it. In scenarios like human robot interaction via natural language, where language input evolves over time, an incremental processing mode would allow an artificial system to behave in a similar way. This includes faster responses due to earlier start of processing and more collaborative behavior by giving timely feedback to the speaker on processing status and established references.
The goal of my project is to design and implement a natural language processing system that provides
an incremental processing mode. The system needs to be able to process partial input and  integrate new input into the previous analysis, if possible, or reanalyse otherwise. Partial analyses will be augmented by anticipations generated from linguistic, semantic and context knowledge.
Another goal is to integrate visual context into the incremental NLP by matching partial analyses with reference candidates from the context. This approach might allow the system to play a more active role in reference resolution, e.g. by guiding visual attention or giving feedback to the speaker.


Cooperations

Multi-modal context integration into language processing is the research topic of two other
Cinacs PhD students, McCrae, Patrick and Baumgärtner,Christopher, with whom I
cooperate. In his project “A Model for the influence of Cross-Modal Context upon Syntactic
Parsing”, P. McCrae designed a system for disambiguating syntactic ambiguities via
semantic knowledge gained from visual context. In my project I am, expanding on this
foundation in cooperation with C. Baumgärtner to permit online context integration during
incremental language processing.

 

General Research Interests

  • Incremental Language Processing
  • Dependency Parsing
  • Psycholinguistically motivated NLP
  • Information Extraction

Presentation of the dissertation project and the research results to date

  
  
Last Updated ( Wednesday, 03 April 2013 )