This week’s natlang lab meeting is on Thursday January 27, 2011 at 2:30pm, in TASC1 9408.
Presentation: Visual analysis of text data by exploiting semantic role labeling (Maxim Roy and Anoop Sarkar)
We will discuss the use of semantic role labeling (SRL) for visual text analytics. The process of navigating large amounts of text data using visualization tools has been tackled by a system called CZSaw built at SIAT in SFU Surrey by John Dill and Rob Woodbury’s research group. One central idea has been that the annotation process that is used to produce an interactive visualization of the data can itself be viewed as a program that can be stored and re-run to produce the same results. We show how visualizations can now be produced based on semantic information embedded in the text.
Readings for discussion are two papers on dual decomposition from EMNLP 2010, the first of which is the EMNLP 2010 best paper:
Terry Koo; Alexander M. Rush; Michael Collins; Tommi Jaakkola; David Sontag. Dual Decomposition for Parsing with Non-Projective Head Automata
Alexander M Rush; David Sontag; Michael Collins; Tommi Jaakkola. On Dual Decomposition and Linear Programming Relaxations for Natural Language Processing