Dataset accompanying:

Fisun, Roman (2026) Zusammengesetzte Indefinitpronomen in den slavischen Sprachen. University of Regensburg

Version

1.0

Description

This dataset contains the semantic-function matrix, the final cluster assignment, and the R code used for the hierarchical cluster analysis of 51 compound indefinite pronoun series presented in Chapter 11.

Contents

funktion.xlsx
    Series-by-function frequency matrix used as input for the cluster analysis.

cluster_membership.csv
    Final six-cluster solution (A–F) for the 51 series.

cluster_functions.R
    R script for the hierarchical cluster analysis. The script performs the main analysis using Euclidean distance and Ward.D2 clustering and a robustness analysis using Spearman distance and average-linkage clustering. It also calculates mean silhouette widths and the Adjusted Rand Index.

Data structure

- Rows in the Excel table represent semantic functions.
- The first column contains the semantic-function labels.
- The remaining columns represent the 51 compound indefinite pronoun series.
- Cells contain non-negative function frequencies.

cluster_membership.csv contains four columns:

- group: cluster label (A–F);
- language: language code;
- modifier: series label;
- group_size: number of series in the respective cluster.

Usage

Place cluster_functions.R and funktion.xlsx in the same directory.

Set this directory as the R working directory and run:

source("cluster_functions.R")

Required R packages:

- readxl
- cluster
- mclust

Project website

https://fisun.org/ZIPs

Author

Roman Fisun
University of Regensburg
https://fisun.org

License

CC BY 4.0

Contact

roman@fisun.org