Semi-Supervised Never-Ending Learning in Rhetorical Relation Identification. Maziero, E. G.; Hirst, G.; and Pardo, T. A. S. In Recent Advances in Natural Language Processing (RANLP-2015), pages ???--???, Hissar, Bulgaria, September, 2015.
abstract   bibtex   
Some languages do not have enough labeled data to obtain good discourse parsing, specially in the relation identification step, and the additional use of unlabeled data is a plausible solution. A workflow is presented that uses a semi-supervised learning approach. Instead of only a pre-defined additional set of unlabeled data, texts obtained from the web are continuously added. This obtains near human perfomance (0.79) in intra-sentential rhetorical relation identification. An experiment for English also shows improvement in accuracy using a similar workflow.
@inproceedings{Maziero2015RANLP,
   author = {Erick Galani Maziero and Graeme Hirst and Thiago A. S. Pardo},
   title = {Semi-Supervised Never-Ending Learning in Rhetorical Relation Identification},
   address = {Hissar, Bulgaria},
   booktitle = {Recent Advances in Natural Language Processing (RANLP-2015)},
   pages = {???--???},
   year = {2015},
   month = {September},
   download = {http://ftp.cs.toronto.edu/pub/gh/Maziero-etal-2015-RANLP.pdf},
   abstract = {Some languages do not have enough labeled data to
                  obtain good discourse parsing, specially in the
                  relation identification step, and the additional use
                  of unlabeled data is a plausible solution. A
                  workflow is presented that uses a semi-supervised
                  learning approach. Instead of only a pre-defined
                  additional set of unlabeled data, texts obtained
                  from the web are continuously added. This obtains
                  near human perfomance (0.79) in intra-sentential
                  rhetorical relation identification. An experiment
                  for English also shows improvement in accuracy using
                  a similar workflow. }
}
Downloads: 0