We propose a complete probabilistic discriminative framework for performing sentencelevel discourse analysis. Our framework comprises a discourse segmenter, based on a binary classifier, and a discourse parser, which applies an optimal CKY-like parsing algorithm to probabilities inferred from a Dynamic Conditional Random Field. We show on two corpora that our approach outperforms the state-of-the-art, often by a wide margin.

@inproceedings{joty-carenini-ng-emnlp-12, abstract = {We propose a complete probabilistic discriminative framework for performing sentencelevel discourse analysis. Our framework comprises a discourse segmenter, based on a binary classifier, and a discourse parser, which applies an optimal CKY-like parsing algorithm to probabilities inferred from a Dynamic Conditional Random Field. We show on two corpora that our approach outperforms the state-of-the-art, often by a wide margin.}, address = {Jeju Island, Korea}, author = {Joty, Shafiq and Carenini, Giuseppe and Ng, Raymond}, booktitle = {Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning}, pages = {904--915}, publisher = {ACL}, series = {EMNLP-CoNLL'12}, title = {A Novel Discriminative Framework for Sentence-Level Discourse Analysis }, url = {http://www.aclweb.org/anthology/D12-1083}, year = {2012} }