Event-Related Features in Feedforward Neural Networks Contribute to Identifying Causal Relations in Discourse

Edoardo Maria Ponti, Anna Korhonen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract / Description of output

Causal relations play a key role in information extraction and reasoning. Most of the times, their expression is ambiguous or implicit, i.e. without signals in the text. This makes their identification challenging. We aim to improve their identification by implementing a Feedforward Neural Network with a novel set of features for this task. In particular, these are based on the position of event mentions and the semantics of events and participants. The resulting classifier outperforms strong baselines on two datasets (the Penn Discourse Treebank and the CSTNews corpus) annotated with different schemes and containing examples in two languages, English and Portuguese. This result demonstrates the importance of events for identifying discourse relations.
Original languageEnglish
Title of host publicationProceedings of the 2nd Workshop on Linking Models of Lexical, Sentential and Discourse-level Semantics
EditorsMichael Roth, Nasrin Mostafazadeh, Nathanael Chambers, Annie Louis
Place of PublicationStroudsburg, PA, USA
PublisherAssociation for Computational Linguistics
Pages25-30
Number of pages6
ISBN (Electronic)978-1-945626-40-1
DOIs
Publication statusPublished - 2 Apr 2017
Event2nd Workshop on Linking Models of Lexical, Sentential and Discourse-level Semantics - Valencia, Spain
Duration: 3 Apr 20173 Apr 2017
http://www.coli.uni-saarland.de/~mroth/LSDSem/

Conference

Conference2nd Workshop on Linking Models of Lexical, Sentential and Discourse-level Semantics
Abbreviated titleLSDSem 2017
Country/TerritorySpain
CityValencia
Period3/04/173/04/17
Internet address

Fingerprint

Dive into the research topics of 'Event-Related Features in Feedforward Neural Networks Contribute to Identifying Causal Relations in Discourse'. Together they form a unique fingerprint.

Cite this