Rich Parsing Workshop

A workshop organized in conjunction with the PhD defense of Andreas van Cranenburgh, for the dissertation entitled "Rich Statistical Parsing and Literary Language." Rich statistical parsing refers here to the use of richer grammar formalisms or techniques; in the thesis these are employed to develop a parser that takes into account discontinuous constituents and produces function tags in addition to the usual syntactic categories. The aim of the workshop is to explore such extensions of (statistical) parsing.

A demo of the parser developed in the context of the thesis is available here: https://lang.science.uva.nl/parser/

Topics:

Location: Room F3.20, enter via Science Park 105 [NikHef], Amsterdam

Date: 1 November 2016

Organizers: Andreas van Cranenburgh, Jelle Zuidema

(preliminary) program:

Time Title Speaker
09h00-09h45 Formal versus statistical enrichment of grammars Andreas van Cranenburgh (ILLC, HHU)
09h45-10h30 LR parsing for Linear Context-Free Rewriting Systems Laura Kallmeyer (HHU)
10h30-11h00 coffee break
11h00-11h45 Minimalist Grammar Transition-Based Parsing Milos Stanojevic (ILLC)
11h45-12h30 Neural syntactic parsing:
gating, embedding, and binding in neural nets
Jelle Zuidema (ILLC)
12h30-14h00 lunch
14h00-14h45 Disambiguation & selectional restrictions Gertjan van Noord (RUG)
14h45-15h30 Strongly lexicalized parsing Tejaswini Deoskar (ILLC)
15h30-16h00 General discussion all speakers

Colocated event: computational linguistics seminar - special session. Same room

Time Title Speaker
16h15-17h00 Distinguishing Figurative and Literal Usages in Discourse Caroline Sporleder
17h00-17h45 Parallel corpora and semantic typology Barend Beekhuizen (joint work with Suzanne Stevenson)

Thesis defense: 2 November 2016, 13:00, Aula der Universiteit (Spui)
The PDF of the thesis is available at https://www.illc.uva.nl/Research/Publications/Dissertations/