%0 Conference Paper %1 schulz2016learning %A Schulz, Sarah %A Kuhn, Jonas %B Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC 2016) %C Slovenia %D 2016 %E Chair), Nicoletta Calzolari (Conference %E Choukri, Khalid %E Declerck, Thierry %E Grobelnik, Marko %E Maegaard, Bente %E Mariani, Joseph %E Moreno, Asuncion %E Odijk, Jan %E Piperidis, Stelios %I European Language Resources Association (ELRA) %K Historical PoS Tagging Text languages low-resource myown %T Learning from Within? Comparing PoS Tagging Approaches for Historical Text %X In this paper, we investigate unsupervised and semi-supervised methods for part-of-speech (PoS) tagging in the context of historical German text. We locate our research in the context of Digital Humanities where the non-canonical nature of text causes issues facing an Natural Language Processing world in which tools are mainly trained on standard data. Data deviating from the norm requires tools adjusted to this data. We explore to which extend the availability of such training material and resources related to it influences the accuracy of PoS tagging. We investigate a variety of algorithms including neural nets, conditional random fields and self-learning techniques in order to find the best-fitted approach to tackle data sparsity. Although methods using resources from related languages outperform weakly supervised methods using just a few training examples, we can still reach a promising accuracy with methods abstaining additional resources.

@inproceedings{schulz2016learning, abstract = {In this paper, we investigate unsupervised and semi-supervised methods for part-of-speech (PoS) tagging in the context of historical German text. We locate our research in the context of Digital Humanities where the non-canonical nature of text causes issues facing an Natural Language Processing world in which tools are mainly trained on standard data. Data deviating from the norm requires tools adjusted to this data. We explore to which extend the availability of such training material and resources related to it influences the accuracy of PoS tagging. We investigate a variety of algorithms including neural nets, conditional random fields and self-learning techniques in order to find the best-fitted approach to tackle data sparsity. Although methods using resources from related languages outperform weakly supervised methods using just a few training examples, we can still reach a promising accuracy with methods abstaining additional resources.}, added-at = {2016-06-08T13:12:46.000+0200}, address = {Slovenia}, author = {Schulz, Sarah and Kuhn, Jonas}, biburl = {https://puma.ub.uni-stuttgart.de/bibtex/2d8e04f51ca5cc496f20cf534ef8ebe4d/sarahschulz}, booktitle = {Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC 2016)}, editor = {Chair), Nicoletta Calzolari (Conference and Choukri, Khalid and Declerck, Thierry and Grobelnik, Marko and Maegaard, Bente and Mariani, Joseph and Moreno, Asuncion and Odijk, Jan and Piperidis, Stelios}, interhash = {305cbe7b196649541c3a42af40a09625}, intrahash = {d8e04f51ca5cc496f20cf534ef8ebe4d}, keywords = {Historical PoS Tagging Text languages low-resource myown}, month = may, publisher = {European Language Resources Association (ELRA)}, timestamp = {2016-06-15T13:28:54.000+0200}, title = {Learning from Within? Comparing PoS Tagging Approaches for Historical Text}, year = 2016 }