Paper: Towards the Unsupervised Acquisition of Discourse Relations

ACL ID P12-2042
Title Towards the Unsupervised Acquisition of Discourse Relations
Venue Annual Meeting of the Association of Computational Linguistics
Session Short Paper
Year 2012

This paper describes a novel approach towards the empirical approximation of discourse re- lations between different utterances in texts. Following the idea that every pair of events comes with preferences regarding the range and frequency of discourse relations connect- ing both parts, the paper investigates whether these preferences are manifested in the distri- bution of relation words (that serve to signal these relations). Experiments on two large-scale English web corpora show that significant correlations be- tween pairs of adjacent events and relation words exist, that they are reproducible on dif- ferent data sets, and for three relation words, that their distribution corresponds to theory- based assumptions. 1 Motivation Texts are not merely accumulations of isolated ut- terances...