IMS at the CoNLL 2017 UD Shared Task: CRFs and Perceptrons Meet Neural Networks

This paper presents the IMS contribution to the CoNLL 2017 Shared Task. In the preprocessing step we employed a CRFPOS/morphological tagger and a neural tagger predicting supertags. On some languages, we also applied word segmenta- tion with the CRF tagger and sentence segmentation with a perceptron-based parser. For parsing we took an ensemble approach by blending multiple instances of three parsers with very different architectures. Our system achieved the third place overall and the second place for the surprise languages

[1]  Joakim Nivre,et al.  Non-Deterministic Oracles for Unrestricted Non-Projective Transition-Based Dependency Parsing , 2015, IWPT.

[2]  Bernd Bohnet,et al.  Top Accuracy and Fast Dependency Parsing is not a Contradiction , 2010, COLING.

[3]  Martin Potthast,et al.  CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies , 2018, CoNLL.

[4]  Yuji Matsumoto,et al.  Improving Dependency Parsers with Supertags , 2014, EACL.

[5]  Jason Eisner Bilexical Grammars and a Cubic-time Probabilistic Parser , 1997, IWPT.

[6]  Agnieszka Falenska,et al.  Stacking or Supertagging for Dependency Parsing - What's the Difference? , 2015, IWPT.

[7]  Mark Steedman,et al.  Using CCG categories to improve Hindi dependency parsing , 2013, ACL.

[8]  Benno Stein,et al.  Improving the Reproducibility of PAN's Shared Tasks: - Plagiarism Detection, Author Identification, and Author Profiling , 2014, CLEF.

[9]  Alon Lavie,et al.  Parser Combination by Reparsing , 2006, NAACL.

[10]  Xavier Carreras,et al.  Experiments with a Higher-Order Projective Dependency Parser , 2007, EMNLP.

[11]  Noah A. Smith,et al.  Turning on the Turbo: Fast Third-Order Non-Projective Turbo Parsers , 2013, ACL.

[12]  Joakim Nivre,et al.  Non-Projective Dependency Parsing in Expected Linear Time , 2009, ACL.

[13]  Ngoc Thang Vu,et al.  Character Composition Model with Convolutional Neural Networks for Dependency Parsing on Morphologically Rich Languages , 2017, ACL.

[14]  Rudolf Rosa,et al.  KLcpos3 - a Language Similarity Measure for Delexicalized Parser Transfer , 2015, ACL.

[15]  Joakim Nivre,et al.  An Improved Oracle for Dependency Parsing with Online Reordering , 2009, IWPT.

[16]  Philip Resnik,et al.  Cross-Language Parser Adaptation between Related Languages , 2008, IJCNLP.

[17]  Ngoc Thang Vu,et al.  A General-Purpose Tagger with Convolutional Neural Networks , 2017, SWCN@EMNLP.

[18]  Hinrich Schütze,et al.  Efficient Higher-Order CRFs for Morphological Tagging , 2013, EMNLP.

[19]  Stephen Clark,et al.  A Tale of Two Parsers: Investigating and Combining Graph-based and Transition-based Dependency Parsing , 2008, EMNLP.

[20]  Jan Hajic,et al.  UDPipe: Trainable Pipeline for Processing CoNLL-U Files Performing Tokenization, Morphological Analysis, POS Tagging and Parsing , 2016, LREC.

[21]  Yuji Matsumoto,et al.  Universal Dependencies 2.0 – CoNLL 2017 Shared Task Development and Test Data , 2017 .

[22]  Jonas Kuhn,et al.  How to Train Dependency Parsers with Inexact Search for Joint Sentence Boundary Detection and Parsing of Entire Documents , 2016, ACL.

[23]  Mihai Surdeanu,et al.  Ensemble Models for Dependency Parsing: Cheap and Good? , 2010, HLT-NAACL.

[24]  Yang Guo,et al.  Structured Perceptron with Inexact Search , 2012, NAACL.

[25]  Fernando Pereira,et al.  Online Learning of Approximate Dependency Parsing Algorithms , 2006, EACL.

[26]  Jonas Kuhn,et al.  Learning Structured Perceptrons for Coreference Resolution with Latent Antecedents and Non-local Features , 2014, ACL.

[27]  Mark Steedman,et al.  Improving Dependency Parsers using Combinatory Categorial Grammar , 2014, EACL.

[28]  Srinivas Bangalore,et al.  The Institute For Research In Cognitive Science Disambiguation of Super Parts of Speech ( or Supertags ) : Almost Parsing by Aravind , 1995 .

[29]  Yuan Zhang,et al.  Stack-propagation: Improved Representation Learning for Syntax , 2016, ACL.

[30]  Nizar Habash,et al.  CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies , 2017, CoNLL.