Organized by Piek Vossen (Vrije Universiteit Amsterdam) and Tommaso Caselli (CLCG, Rijksuniversiteit Groningen).
This workshop will be organised in the form of tutorials and hands-on data sessions in order to make participant familiar with methodologies in the area of Natural Language Processing and Computational Linguistics for the annotation, both manual and automatic, and visualisation of storylines. Storylines are defined as structured index of events, coherently ordered.
The workshop will be organised around three major topics/working areas: i.) introduction to the storyline framework; ii.) annotation of storylines: annotation schemas and tools; iii.) automatic ex- traction and visualisation of storylines.
The storyline framework: this session will be mainly dedicated to setting the common grounds by presenting the storyline framework, its main concepts (events, temporal expression, temporal relations, and explanatory relations), the corresponding computational model.
Annotation of Storylines: this session will be divided into two parts: the first part will be dedicat- ed to the presentation of two annotation schemes (the ECBStar annotation scheme and the Con- tent Types Annotation Scheme) and tools (Content Annotation Tool - CAT). The second part will be dedicated to hands-on data. Participants will apply the schemas to a dataset (their own or a dataset provided by the organisers). This exercise will lead to discussions and reflections on re- quirements, adaptations of existing schemas, or the development of new annotation schemes to be integrated in the storyline framework.
Extraction and Visualisation of Storylines: this session will show how the available computa- tional tools for the extraction and visualisation of storylines developed in the NewsReader and the Spinoza-ULM 3 projects.
Bal, Mieke. Narratology: Introduction to the theory of narrative. University of Toronto Press, 2009.
Caselli, Tommaso, and Piek Vossen. "The Storyline Annotation and Representation Scheme (StaR): A Proposal." Proceedings of the 2nd Workshop on Computing News Storylines (CNS 2016). 2016.
Caselli, Tommaso, and Piek Vossen. "The Event StoryLine Corpus: A New Benchmark for Causal and Temporal Relation Extraction." Proceedings of the Events and Stories in the News Workshop (EventStory 2017). 2017
Lenzi, Valentina Bartalesi, Giovanni Moretti, and Rachele Sprugnoli. "CAT: the CELCT Annotation Tool." LREC. 2012.
van Meersbergen, Maarten, et al. "Storyteller: Visual analytics of perspectives on rich text inter- pretations." Proceedings of the 2017 EMNLP Workshop: Natural Language Processing meets Journalism. 2017.
Pustejovsky, James, et al. "ISO-TimeML: An International Standard for Semantic Annotation." LREC. 2010.
Segers, Roxane, Tommaso Caselli, and Piek Vossen. "The Circumstantial Event Ontology (CEO)." Proceedings of the Events and Stories in the News Workshop (EventStory 2017). 2017.
Sprugnoli, Rachele, and Sara Tonelli. "One, no one and one hundred thousand events: Defining and processing events in an inter-disciplinary perspective." Natural Language Engineering 23.4 (2017): 485-506.
Van Hage, Willem Robert, et al. "Design and use of the Simple Event Model (SEM)." Web Seman- tics: Science, Services and Agents on the World Wide Web 9.2 (2011): 128-136.
Vossen, Piek, Tommaso Caselli, and Yiota Kontzopoulou. "Storylines for structuring massive streams of news." Proceedings of the First Workshop on Computing News Storylines. 2015.
Vossen, Piek, et al. "Newsreader: Using knowledge resources in a cross-lingual reading machine to generate more knowledge from massive streams of news." Knowledge-Based Systems 110 (2016): 60-85.
Dr. Tommaso Caselli (see picture below) is Assistant Professor at the University of Groningen. He has been involved in research in NLP since 2006. His main research areas are in event processing, temporal relations and (event) sentiment analysis. His research has recently focused on developing computational models and NLP tools for semantic processing of texts with a particular focus on cross-document event extraction and storyline extraction from news. He took part to the organisation of semantic evaluation campaigns in NLP for English and Italian (SemEval 2010 TempEval-2; EVALITA 2014 EVENTI, SemEval 2015 CLIPEval, EVALITA 2016 FactA). He is one of the initiator of the Comput- ing News Storylines workshop series, now at its fourth edition. He received his MA in Linguistics from the University of Pavia, and his Ph.D. in Computational Linguistics from the University of Pisa. Before joining the University of Groningen, he was a post-doc at the Institute of Computational Linguistics in Pisa (2009-2011), Hong Kong Polytechnic Univeristy (2011-2012), IBM Center for Advanced Studies in Trento (2012-2014), and at the Vrije Universiteit Amsterdam (2014-2017).
Prof. Dr. Piek Vossen (see picture below) received his PhD (Cum Laude) in 1995 on Computational Lexicology. He is now professor at the Faculty of Humanities of the VU University of Amsterdam and the head of CLTL (Computational Lexicology and Terminology Lab). He coordinated various EU projects: Eu- roWordNet-I-II (LE24003 and LE8328), KYOTO (FP7, ICT-211423) and most recently NewsReader (ICT-316404) and participated in many others among which Acquilex-I-II (BRA7315 and BRA3030), SIFT (LRE62030), MEANING (IST-2001-34460), OpeNER (ICT-296451). In his position at the VU, he also led a number of NWO projects in the area of lexicology among which Cornetto and DutchSemCor. In 2013, he received the prestigious Spinoza-prize for his groundbreaking re- search on wordnets and NewsReader. He also received the Enlighten-Your- Research prize in 2013 for NewsReader’s challenge to process daily news-streams of millions of articles. Recently, he is extending this research into social robots that communicate about the world around us. Vossen is the co-founder and co-president of the Global-Wordnet-Association that organized 9 international conferences since 2002. He is also involved in the national infrastructure projects CLARIN-NL and CLARIAH.