TERENCE
An Adaptive Learning System for Reasoning about Stories with Poor Comprehenders and their Educators
Aims
TERENCE combines natural language processing techniques with insights from educational psychology to develop an adaptative learning system for 8-10 year old children with text comprehension difficulties. TERENCE employs English and Italian stories for poor comprehenders as reading material, and tackles high-level cognitive text processing through smart games that improve reasoning skills. Natural language processing models take the text of the stories and automatically identify the important events and the temporal and causal relations between them to build up a semantic representation of the narrative. Smart games are then constructed on top of these semantic representations to stimulate children to reason about the story events, thereby improving their literacy.
In this project, LIIR leads the research on event, temporal and causal information extraction in stories.
Partners
The other TERENCE partners are the University of Aquila, Italy, AMNIN, Slovenia, the Centre for Biomedicine, Intelligent Systems, Educational, Technology, Spain, the FBK Research Centre, Italy, the Free University of Bozen-Bolzano, Italy, the L3S Research Centre, Hannover, Germany, Moholy-Nagy University in Budapest, Hungary, the University of Padova, Italy, SIVECO Romania, the University of Sussex, Great Britain, and the University of Verona, Italy. LIIR closely collaborates with the Free University of Bozen-Bolzano (Prof. Rosella Gennari) and with the the L3S Research Centre, Hannover (Prof. Wolfgang Nejdl).
.Results
We have developed technology for the recognition of events. We have built a novel model that considers temporal relation extraction as a form of dependency parsing, and have successfully evaluated this model on English stories. The research has already resulted in a first demonstrator that recognizes the events, their actors and coreferents, and temporal relations in a story (realized by of Oleksandr Kolomiyets and Jan De Belder).
| Period | From 2010-10-01 to 2013-09-30. |
| Financed by | EU FP7- 257410 |
| Supervised by | Marie-Francine Moens |
| Staff | Steven Bethard |
| Nicla Civitella | |
| Jan De Belder | |
| Oleksandr Kolomiyets | |
| Contact | Oleksandr Kolomiyets |
| More information can be found on the project website http://www.terenceproject.eu/ | |
Publications
- KOLOMIYETS, Oleksandr, BETHARD, Steven & MOENS, Marie-Francine Model-Portability Experiments for Textual Temporal Analysis. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies (pp. 271-276). ACL. 2011
- JANS, Bram, VULIC, Ivan, BETHARD, Steven & MOENS, Marie-Francine Skip N-grams and Ranking Functions for Predicting Script Events. In Proceedings of the Thirteenth Conference of the European Chapter of the Association for Computational Linguistics (EACL 2012). ACL. 2012
- KOLOMIYETS, Oleksandr, BETHARD, Steven & MOENS, Marie-Francine Extracting Narrative Timelines as Temporal Dependency Structures. In Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics (pp. 87-98). ACL 2012
- BETHARD, Steven, KOLOMIYETS, O. & MOENS, Marie-Francine Annotating Story Timelines as Temporal Dependency Structures. In Proceedings of the International Conference on Language Resources and Evaluation Conference (LREC 2012) . ELRA. 2012
- DE BELDER, Jan & MOENS, Marie-Francine Coreference Clustering using Column Generation. In Proceedings of the 24th International Conference on Computational Linguistics. ACL. 2012
Back to all projects
