Information distance reveals when and by whom literary texts were written

This deposit contains the data and code needed to reproduce the results of the paper "Information distance reveals when and by whom literary texts were written" by A. Mesa-Rodríguez, D. Estevez-Moya, E. Estevez-Rams and H. Kantz, and of its SI Appendix. The study uses a Lempel–Ziv estimate of the information distance to date English literary works and attribute their authorship, and to locate where the signals of period and author reside in a text. The corpus consists of 945 public-domain Project Gutenberg texts by 157 authors born between 1415 and 1896. The deposit includes the cleaned texts, the five representations analysed in the paper (the letter sequence, the sentence-shuffled and word-shuffled texts, the function words and the punctuation) with the control representations of the SI, the 945 × 945 matrices of the information distance and of the comparison distances (bzip2 compression distance, Burrows' Delta and character n-grams), the pretrained text embeddings (all-MiniLM-L6-v2, LUAR and MacBERTh), the metadata of every text with Datasets S1 to S4, and the Python scripts that build the representations and run every analysis. Information distances were computed with the lzcomplexity package (https://doi.org/10.5281/zenodo.23531093). The README describes each file and how to run the scripts. The data are released under CC BY 4.0 and the code under the MIT license.