Shallow Discourse Parsing with Conditional Random Fields

Parsing discourse is a challenging natural language processing task. In this paper we take a data driven approach to identify arguments of explicit discourse connectives. In contrast to previous work we do not make any assumptions on the span of arguments and consider parsing as a token-level sequence labeling task. We design the argument segmentation task as a cascade of decisions based on conditional random fields (CRFs). We train the CRFs on lexical, syntactic and semantic features extracted from the Penn Discourse Treebank and evaluate feature combinations on the commonly used test split. We show that the best combination of features includes syntactic and semantic features. The comparative error analysis investigates the performance variability over connective types and argument positions.

[1]  Livio Robaldo,et al.  The Penn Discourse TreeBank 2.0. , 2008, LREC.

[2]  Rich Caruana,et al.  Greedy Attribute Selection , 1994, ICML.

[3]  Jason Baldridge,et al.  Discourse Connective Argument Identification with Connective Specific Rankers , 2008, 2008 IEEE International Conference on Semantic Computing.

[4]  B. D. Ripley.  COMPUTER-INTENSIVE STATISTICAL METHODS , 2011 .

[5]  Ani Nenkova,et al.  Easily Identifiable Discourse Relations , 2008, COLING.

[6]  Richard Johansson,et al.  Syntactic and Semantic Structure for Opinion Expression Detection , 2010, CoNLL.

[7]  Alan Lee,et al.  Attribution and the (Non-)Alignment of Syntactic and Discourse Arguments of Connectives , 2005, FCA@ACL.

[8]  Robert Tibshirani,et al.  Computer‐Intensive Statistical Methods , 2006 .

[9]  James Pustejovsky,et al.  Sequence models and ranking methods for discourse parsing , 2009 .

[10]  Rashmi Prasad,et al.  Exploiting Scope for Shallow Discourse Parsing , 2010, LREC.

[11]  Yuji Matsumoto,et al.  Statistical Dependency Analysis with Support Vector Machines , 2003, IWPT.

[12]  William C. Mann,et al.  Rhetorical Structure Theory: Toward a functional theory of text organization , 1988 .

[13]  Beatrice Santorini,et al.  Building a Large Annotated Corpus of English: The Penn Treebank , 1993, CL.

[14]  Ani Nenkova,et al.  Using Syntax to Disambiguate Explicit Discourse Connectives in Text , 2009, ACL.

[15]  Fernando Pereira,et al.  Shallow Parsing with Conditional Random Fields , 2003, NAACL.

[16]  Pat Langley,et al.  Editorial: On Machine Learning , 1986, Machine Learning.

[17]  Lise Getoor,et al.  Supervised and Unsupervised Methods in Employing Discourse Relations for Improving Opinion Polarity Classification , 2009, EMNLP.

[18]  Rashmi Prasad,et al.  Annotation of Discourse Relations for Conversational Spoken Dialogs , 2010, LREC.

[19]  James Pustejovsky,et al.  Automatically Identifying the Arguments of Discourse Connectives , 2007, EMNLP.

[20]  Bonnie L. Webber,et al.  Discourse structure and language technology , 2011, Natural Language Engineering.

[21]  Ani Nenkova,et al.  Automatic sense prediction for implicit discourse relations in text , 2009, ACL.

[22]  Hwee Tou Ng,et al.  Recognizing Implicit Discourse Relations in the Penn Discourse Treebank , 2009, EMNLP.

[23]  Andrew McCallum,et al.  Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data , 2001, ICML.

[24]  John A. Carroll,et al.  Applied morphological processing of English , 2001, Natural Language Engineering.

[25]  Richard Johansson,et al.  End-to-End Discourse Parser Evaluation , 2011, 2011 IEEE Fifth International Conference on Semantic Computing.