Automatic extraction of causal knowledge from natural language texts

Diplôme Thèse Automatic extraction of causal knowledge from natural language texts eng 2012-05-16 author Grivaz Cécile 635820 director Moeschler Jacques 26402 director Rajmann Martin 542440 10.13097/archive-ouverte/unige:24660 urn:nbn:ch:unige-246603 L. 747 This thesis studies the automatic recognition of implicit causal relations between clauses. Previous work, although more accurate than a random baseline, does not achieve sufficient accuracy for practical use. First we show, using annotation experiments, that recognising implicit causation is a subjective task, in spite of its association with several observable features. We then propose an evaluation protocol that takes the subjectivity of the task into account. Previous linguistics work uses the notion of world knowledge. We show that the most likely feature for representing this knowledge -verb pairs- is not predictive of causation in practice. We then show that the current state of the art is not sufficient to allow us to represent nor to acquire the world knowledge that is necessary for this task. We conclude that the field cannot make important progress without first solving the problem of abstract eventuality representation and clustering. Natural language processing Computational linguistics Information extraction Computational semantics Computational pragmatics Discoure relation classification Causation Causal relation recognition Dewey 410 82 Département de linguistique 39 false 3ad93e28-9784-4473-aaf4-711e4d58b7be Thesis Public theseGrivaz.pdf 1647982 application/pdf openAccess Faculté des lettres / Département de linguistique

Title	Published in	Access level	OA Policy	Year	Views	Downloads

Automatic extraction of causal knowledge from natural language texts				2012	1,143	1,599

Archive ouverte UNIGE

Grivaz, Cécile