Summary of the paper
| Title | Human Judgements on Causation in French Texts |
|---|---|
| Authors | Cécile Grivaz |
| Abstract | The annotation of causal relations in natural language texts can lead to a lowinter-annotator agreement. A French corpus annotated with causal relationswould be helpful for the evaluation of programs that extract causal knowledge,as well as for the study of the expression of causation. As previoustheoretical work provides no necessary and sufficient condition that wouldallow an annotator to easily identify causation, we explore features that areassociated with causation in human judgements. We present an experiment thatallows us to elicit intuitive features of causation. We test the statisticalassociation of features of causation from theoretical previous work withcausation itself in human judgements in an annotation experiment. We thenestablish guidelines based on these features for annotating a French corpus. Weargue that our approach leads to coherent annotation guidelines, since itallows us to obtain a κ = 0.84 agreement between the majority of theannotators answers and our own educated judgements. We present these annotationinstructions in detail. |
| Language | Question Answering |
| Topics | Corpus (creation, annotation, etc.), Information Extraction, Information Retrieval, Question Answering |
Full paper  |
Human Judgements on Causation in French Texts |
| Bibtex | @InProceedings{GRIVAZ10.145, author = {Cécile Grivaz}, title = {Human Judgements on Causation in French Texts}, booktitle = {Proceedings of the Seventh conference on International Language Resources and Evaluation (LREC'10)}, year = {2010}, month = {may}, date = {19-21}, address = {Valletta, Malta}, editor = {Nicoletta Calzolari (Conference Chair), Khalid Choukri, Bente Maegaard, Joseph Mariani, Jan Odjik, Stelios Piperidis, Mike Rosner, Daniel Tapias}, publisher = {European Language Resources Association (ELRA)}, isbn = {2-9517408-6-7}, language = {english} } |