TY  - CHAP
U1  - Buchbeitrag
A1  - Hitschler, Julian
A1  - van den Berg, Esther
A1  - Rehbein, Ines
T1  - Authorship attribution with convolutional neural networks and POS-eliding
T2  - Proceedings of the Workshop on Stylistic Variation (EMNLP 2017). September 8, 2017 Copenhagen, Denmark
N2  - We use a convolutional neural network to perform authorship identification on a very homogeneous dataset of scientific publications. In order to investigate the effect of domain biases, we obscure words below a certain frequency threshold, retaining only their POS-tags. This procedure improves test performance due to better generalization on unseen data. Using our method, we are able to predict the authors of scientific publications in the same discipline at levels well above chance.
KW  - Autorschaft
KW  - Computerlinguistik
KW  - Part-of-Speech-Tagging
Y1  - 2017
UN  - https://nbn-resolving.org/urn:nbn:de:bsz:mh39-80252
UR  - http://aclweb.org/anthology/W17-4907
SN  - 978-1-945626-99-9
SB  - 978-1-945626-99-9
U6  - https://doi.org/10.18653/v1/W17-4907
DO  - https://doi.org/10.18653/v1/W17-4907
SP  - 53
EP  - 28
PB  - The Association for Computational Linguistics
CY  - Stroudsburg PA, USA
ER  -