Section: New Results

Discourse Parsing

Participants : Laurence Danlos, Chloé Braud.

Discourse parsing goal is to reflect the rhetorical structure of a document, how pieces of text are linked in order to form a coherent document. Understanding such links could benefits to several other natural language applications (summarization, language generation, information extraction...). A discourse parser corresponds to two major subtasks: a segmentation step wherein discourse units (DUs) are extracted, and a parsing step wherein these DUs are (recursively) related through “discourse (rhetorical) relations”. The more difficult task in discourse parsing is the labeling of the relations between DUs, especially when no so-called connective overtly marks the relation (we then talk about implicit relations as opposed to explicit ones). In her PhD work, Chloé Braud develops a discourse relation classifier, carrying experiments on French and English. Focusing on the problem on implicit relation identification, this work tries to tackle the lack of manually annotated data, a discourse specific difficulty, by exploiting the similarities between explicit and implicit relations. In 2014, this work lead to systems based on domain adaptation methods [18] , [13] , demonstrating improvements on the French corpus Annodis [56] .