Datasets and Source Code

Research datasets and source code tied to individual publications are listed below. All data is made available for non-commercial scientific research only.

Sentiment and Linguistic Features

R code and dataset · Schneider (2026)

Dataset and R code accompanying the 2026 study "Pop Lyrics Through Time: Challenges in Corpus-Based Modeling of Linguistic and Emotional Dynamics in German Pop Lyrics".

Dataset and R-Code ↓

Oral–Literate Continuum Classifiers

Random Forest classifiers (R RDS format) · Broll & Schneider (2023)

Pre-trained Random Forest classifiers for positioning song texts on the oral–literate continuum, as described in the 2023 study by Broll and Schneider.

Random Forest Classifiers in R Data Format (RDS) ↓

Idiomatic Language Dataset

Dataset and Jupyter Notebook · Amin et al. (2021)

Dataset and annotation notebook from the 2021 MWE Workshop publication on data-driven identification of idioms in song lyrics.

Dataset and Jupyter Notebook pipeline ↓

Sociopolitical Vocabulary

Keyword list and distributional data · Schneider et al. (2022)

Keyword list and distributional corpus data from the 2022 study examining sociopolitical vocabulary in German song lyrics across five decades.

Sociopolitical keyword list and distributional data ↓

Rhyme Detection

Python code and evaluation files · University of Leipzig (2022)

Python implementation and evaluation datasets for automated rhyme detection in German song lyrics.

Python code and evaluation files ↓

Please cite the respective publication when using a dataset; the literature references are clickable. See also Publications.