TY - CHAP U1 - Konferenzveröffentlichung A1 - Brunner, Annelen A1 - Tu, Ngoc Duyen Tanja A1 - Weimer, Lukas A1 - Jannidis, Fotis T1 - Deep learning for free indirect representation T2 - Preliminary proceedings of the 15th Conference on Natural Language Processing (KONVENS 2019), October 9 – 11, 2019 at Friedrich-Alexander-Universität Erlangen-Nürnberg N2 - In this paper, we present our work-inprogress to automatically identify free indirect representation (FI), a type of thought representation used in literary texts. With a deep learning approach using contextual string embeddings, we achieve f1 scores between 0.45 and 0.5 (sentence-based evaluation for the FI category) on two very different German corpora, a clear improvement on earlier attempts for this task. We show how consistently marked direct speech can help in this task. In our evaluation, we also consider human inter-annotator scores and thus address measures of certainty for this difficult phenomenon. KW - Deutsch KW - Indirekte Rede KW - Erlebte Rede KW - Automatische Sprachanalyse KW - Korpus Y1 - 2019 U6 - https://nbn-resolving.org/urn:nbn:de:bsz:mh39-93151 UN - https://nbn-resolving.org/urn:nbn:de:bsz:mh39-93151 UR - https://corpora.linguistik.uni-erlangen.de/data/konvens/proceedings/papers/KONVENS2019_paper_27.pdf SP - 241 EP - 245 PB - German Society for Computational Linguistics & Language Technology und Friedrich-Alexander-Universität Erlangen-Nürnberg CY - München [u.a.] ER -