Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Two Attempts to Predict Author Gender in Cross-Genre Settings in Dutch

 
: Brito, Eduardo; Sifa, Rafet; Bauckhage, Christian

:
Fulltext urn:nbn:de:0011-n-5934648 (363 KByte PDF)
MD5 Fingerprint: 2b90ef273f1b27bc0525f4dde5c37594
Created on: 8.7.2020


Haagsma, Hessel:
Proceedings of the Shared Task on Cross-Genre Gender Prediction in Dutch at GxG@CLIN29. Online resource : Cco-located with the 29th Conference on Computational Linguistics in The Netherlands (CLIN29) Groningen, The Netherlands, January 31, 2019
Online im WWW, 2019 (CEUR Workshop Proceedings 2453)
http://ceur-ws.org/Vol-2453/
URN: urn:nbn:de:0074-2453-4
pp.22-29
Shared Task on Cross-Genre Gender Prediction in Dutch (GxG-CLIN29) <2019, Groningen>
Conference on Computational Linguistics in The Netherlands (CLIN) <29, 2019, Groningen>
English
Conference Paper, Electronic Publication
Fraunhofer IAIS ()

Abstract
This paper describes the systems designed by the FraunhoferIAIS team at the CLIN29 shared task on cross-genre gender detection in Dutch. We show two alternative classification approaches: a rather standard one consisting of feature engineering and a random forest classifier; and an alternative one involving a LSTM classifier. Both are enhanced by a LDA model trained on stems. We considered various features such as frequency of function words, parts-of-speech and sentiment among others. We achieved 53.77% average accuracy in the cross-genre settings.

: http://publica.fraunhofer.de/documents/N-593464.html