TY - CHAP U1 - Konferenzveröffentlichung A1 - Amin, Miriam A1 - Fankhauser, Peter A1 - Kupietz, Marc A1 - Schneider, Roman ED - Cook, Paul ED - Mitrović, Jelena ED - Parra Escartín, Carla ED - Vaidya, Ashwini ED - Osenova, Petya ED - Taslimipoor, Shiva ED - Ramisch, Carlos T1 - Data-driven identification of idioms in song lyrics T2 - Proceedings of the 17th Workshop on Multiword Expressions (MWE 2021) N2 - The automatic recognition of idioms poses a challenging problem for NLP applications. Whereas native speakers can intuitively handle multiword expressions whose compositional meanings are hard to trace back to individual word semantics, there is still ample scope for improvement regarding computational approaches. We assume that idiomatic constructions can be characterized by gradual intensities of semantic non-compositionality, formal fixedness, and unusual usage context, and introduce a number of measures for these characteristics, comprising count-based and predictive collocation measures together with measures of context (un)similarity. We evaluate our approach on a manually labelled gold standard, derived from a corpus of German pop lyrics. To this end, we apply a Random Forest classifier to analyze the individual contribution of features for automatically detecting idioms, and study the trade-off between recall and precision. Finally, we evaluate the classifier on an independent dataset of idioms extracted from a list of Wikipedia idioms, achieving state-of-the art accuracy. KW - Phraseologie KW - Lyrics KW - Automatische Spracherkennung KW - Automatische Sprachanalyse KW - Komposition KW - Semantik KW - Deutsch KW - natural language processing KW - multiword expressions Y1 - 2021 UN - https://nbn-resolving.org/urn:nbn:de:bsz:mh39-106825 SN - 978-1-954085-71-8 SB - 978-1-954085-71-8 U6 - https://doi.org/10.18653/v1/2021.mwe-1.3 DO - https://doi.org/10.18653/v1/2021.mwe-1.3 SP - 13 EP - 22 PB - Association for Computational Linguistics CY - Stroudsburg ER -