Conference Schedule
Thursday, December 13, 2018
09:00–09:15 Registration
09:15–09:30 Opening
09:30–10:30 Keynote by Adam Przepiórkowski: Coordination in Universal Dependencies
10:30–11:00 Coffee Break
11:00–11:30 Gosse Bouma: Comparing two Methods for Adding Enhanced Dependencies to UD treebanks
11:30–12:00 Sonja Markovi? and Daniel Zeman: Reflexives in Universal Dependencies
12:00–12:20 Jirka Hana and Barbora Hladka: Universal Dependencies and a Non-Native Czech
12:20–13:30 Lunch
13:30–14:00 Tatiana Bladier, Andreas van Cranenburgh, Kilian Evang, Laura Kallmeyer, Robin M?llemann and Rainer Osswald: RRGbank: a Role and Reference Grammar Corpus of Syntactic Structures Extracted from the Penn Treebank
14:00–14:30 Dan Flickinger: Measuring Evolution of Implemented Grammars
14:30–15:00 Matías Guzmán Naranjo and Laura Becker: Quantitative Word Order Typology with UD
15:00–15:20 Peter Bourgonje and Manfred Stede: Potsdam Commentary Corpus 2.1 in ANNIS3
15:20–16:00 Coffee Break
16:00–17:30 Panel Discussion:
19:00–21:00 Conference Dinner
Friday, December 14, 2018
09:00–09:30 Eleni Metheniti and Gunter Neumann: Wikinflection: Massive Semi-Supervised Generation of Multilingual Inflectional Corpus from Wiktionary
09:30–10:00 Tobias Pütz, Dani?l de Kok, Sebastian Pütz and Erhard Hinrichs: Sequence2Sequence or Perceptrons for Lemmatization. A Thorough Error Analysis
10:00–10:30 Noor Abo Mokh and Sandra Kübler: Preprocessing Does Matter: Parsing Non-Segmented Arabic
10:30–11:00 Coffee Break
11:00–11:30 Eva Fu?íková, Eva Haji?ová, Jan Haji? and Zdeňka Ure?ová: Defining Verbal Synonyms: between Syntax and Semantics
11:30–12:00 Alastair Butler and Stephen Wright Horn: Parsed Annotation with Semantic Calculation
12:00–13:00 Lunch
13:00–13:30 Sardar Jaf and Thomas G Hudson: On the Development of a Large Scale Corpus for Native Language Identification
13:30–14:00 Atreyee Mukherjee and Sandra Kübler (Presented by Francis Tyers): Domain Adaptation in Dependency Parsing via Transformation Based Error Driven Learning
14:00–14:30 Kira Droganova, Olga Lyashevskaya and Daniel Zeman: Data Conversion and Consistency of Monolingual Corpora: Russian UD Treebanks
14:30–15:00 Coffee Break
15:00–16:00 Keynote by Malvina Nissim: Using Weak Signal in NLP
16:00–16:10 Closing Session (#syntaxfest)
Invited Speakers
Using Weak Signal in NLP
Malvina Nissim (University of Groningen)
Treebanking requires substantial expert labour in the annotation of a variety of language phenomena. Possibly to a lesser extent for phenomena where laypeople can also contribute, the need for assigning manual labels nevertheless characterises almost all language processing tasks, since they are usually best solved by supervised models. Such models are indeed accurate, but we also know that they lack portability, as they are bound to languages, genres, and even specific datasets.
Having spent years dealing with annotation issues and label acquisition for various semantic and pragmatic tasks, in this talk I take a radically different perspective, which hopefully can yield interesting reflections over treebanking, too. I will show various ways to cheaply obtain and exploit weaker signal in supervised learning, even venturing on the suggestion to reduce existing strong, accurate signal in order to enhance portability. I will do so via discussing three case studies in three different classification tasks, all focused on social media.
Coordination in Universal Dependencies
Adam Przepiórkowski (Polish Academy of Sciences and University of Warsaw)
The aim of this talk is to review theoretical dependency approaches to coordination and to propose an extension to the Universal Dependencies (UD) treatment of this phenomenon. Some emphasis will be put on aspects of coordination which are currently problematic for UD, namely, nested coordination and coordination of unlike grammatical functions.
Social Activities
There will be a joint conference dinner on the evening of Thursday, December 13, 2018.