PoeTree_en: English Poetry Treebanks & Annotated Corpus

PoeTree_en is the English subsection of the PoeTree standardized multilingual collection of poetry corpora.
The dataset comprises 40,282 poems written by 430 distinct poets, fully parsed into individual JSON documents with rich metadata, author biographies, book sources, full-text poem lines, and sentence-level Universal Dependencies (CoNLL-U) token annotations (lemmas, UPOS, XPOS, morphological features, dependency parse trees, and phonetic/metrical alignments).

The English poetry subset was derived from the Project Gutenberg open corpus and is entirely within the Open Access Public Domain.

---

Repository Organization

All poems are organized into 430 subfolders titled by the poet's name following the convention LastName_FirstName_MiddleName (with punctuation removed):

PoeTree_en/
├── README.md
├── Abbey_Henry/
├── Adams_John_Quincy/
├── Akenside_Mark/
├── Alcott_Amos_Bronson/
├── Alcott_Louisa_May/
├── Alderson_Matthew_W/
├── Aldrich_Anne_Reeve/
├── Alger_Horatio/
├── Allen_Hervey/
├── Allingham_William/
├── Allston_Washington/
├── Alma-Tadema_Laurence/
├── Arnold_Edwin/
├── Arnold_Matthew/
├── ...
├── Whitman_Walt/
├── Whittier_John_Greenleaf/
├── Wilcox_Ella_Wheeler/
├── Williams_Helen_Maria/
├── Williams_James/
├── Williams_William_Carlos/
├── Wordsworth_William/
├── Wylie_Elinor/
└── Young_Francis_Brett/

Corpus Statistics

• Total Poems: 40,282 JSON files
• Total Authors: 430 poets
• License / Rights: CC BY-SA 4.0 (PoeTree data), Open Access / Public Domain (Project Gutenberg-derived works)
• Format: Standalone JSON per poem (UTF-8)

---

📖 JSON Schema Reference

All corpora in PoeTree follow the exact standardized JSON schema defined at https://versologie.cz/poetree/json-schema/.

Each poem is stored in an individual JSON file with the following attributes:

Top-Level Schema

Field	Type	Description
id	str	Unique ID of the poem (points to a source file).
title	`str	null`
year_created	`int	list
author	`dict	list`
source	dict	Metadata dictionary describing the book/volume source.
body	list[dict]	List of line dictionaries containing verse text, stanzas, and token-level CoNLL-U linguistic annotations.
neighbors	list	20 most similar poems (fuzzy partial ratio) by the same author, encoded as [id, similarity_score] in descending order.
duplicate	`str	false`
locations	`list[dict]	false`

---

Object Definitions

1. author (Author Metadata)

{
  "name": "Whitman, Walt",
  "wiki": "Q81438",
  "viaf": "2478331",
  "born": 1819,
  "died": 1892,
  "country": "us"
}

• name (str): Name of the author as it appears in the source (or [anonymous]).
• viaf (str | null): Virtual International Authority File (VIAF) ID.
• wiki (str | null): Wikidata entity identifier (e.g., Q81438).
• country (str | null): Country of origin (ISO code) or null if unknown.
• born (int | null): Year of birth.
• died (int | null): Year of death.

2. source (Publication Source)

{
  "id": "leavesOfGrass",
  "title": "Leaves of Grass",
  "year_published": 1855,
  "publisher": null,
  "place": "Brooklyn, New York",
  "corpus": "gutenberg"
}

• id (str | null): Book/volume identifier.
• title (str | null): Title of the source book.
• year_published (int | null): Year of publication.
• publisher (str | null): Publisher name.
• place (str | null): Place/city of publication.
• corpus (str): Source corpus name (gutenberg for English).

3. locations (Geolocation Entities)

• wiki (str): Location Wikidata identifier (e.g. Q1085 for Prague).
• loc_start ([line_i, word_i]): Zero-based line index and word index of entity start.
• loc_end ([line_i, word_i]): Zero-based line index and word index of entity end.

4. body (Lines & CoNLL-U Word Annotations)

Each element in body represents a verse line:

{
  "id": 0,
  "stanza_id": 0,
  "text": "I celebrate myself, and sing myself,",
  "part": false,
  "words": [
    {
      "id": 1,
      "form": "I",
      "lemma": "I",
      "upos": "PRON",
      "xpos": "PRP",
      "feats": "Case=Nom|Number=Sing|Person=1|PronType=Prs",
      "head": 2,
      "deprel": "nsubj",
      "deps": "_",
      "sentence": 1
    }
  ]
}

• id (int): Zero-based line index, incrementing monotonically throughout the entire poem (does not reset across stanzas).
• stanza_id (int): Zero-based stanza index.
• text (str): Surface text of the verse line.
• part (str | false): For split/divided lines: "I" (initial), "M" (medial), "F" (final), or false for normal undivided lines.
• words (list[dict]): Word-level linguistic annotations according to the Universal Dependencies (CoNLL-U) standard:
  • id (int): 1-based token index within the sentence.
  • form (str): Surface word form or punctuation symbol.
  • lemma (str): Lemma or stem of the word form.
  • upos (str): Universal Part-of-Speech tag (e.g., NOUN, VERB, ADJ, PRON, PUNCT).
  • xpos (str): Language-specific part-of-speech tag (_ if unavailable).
  • feats (str): Morphological features from universal inventory (_ if unavailable).
  • head (int): Head governor token index (0 if root).
  • deprel (str): Universal dependency relation to the head (e.g., nsubj, obj, root).
  • deps (str): Enhanced dependency graph representation.
  • sentence (int): 1-based sentence index.
  • multiword (dict, optional): Multiword token details if token is part of a compound contraction.

---

How to Cite

If you use PoeTree in your research, please cite the dataset and the primary journal article as reproduced from https://versologie.cz/poetree/cite:

1. PoeTree Dataset

@dataset{poetree_data,
  author    = {Plecháč, Petr and
              Kolár, Robert and
              Cinková, Silvie and
              Šeļa, Artjoms and
              De Sisto, Mirella and
              Nugues, Lara and
              Haider, Thomas and
              Nagy, Benjamin and
              Delente, Éliane and
              Renault, Richard and
              Bobenhausen, Klemens and
              Hammerich, Benjamin and
              Mittmann, Adiel and
              Palkó, Gábor and
              Horváth, Péter and
              Navarro Colorado, Borja and
              Ruiz Fabo, Pablo and
              Bermúdez Sabel, Helena and
              Korchagin, Kirill and
              Plungian, Vladimir and
              Sitchinava, Dmitri},
  title     = {{PoeTree. Poetry Treebanks in Czech, English, 
              French, German, Hungarian, Italian, Portuguese,
              Russian and Spanish}},
  month     = oct,
  year      = 2023,
  publisher = {Zenodo},
  version   = {0.0.1},
  doi       = {10.5281/zenodo.10008459},
  url       = {https://doi.org/10.5281/zenodo.10008459}
}

APA Citation:

Plecháč, P., Kolár, R., Cinková, S., Šeļa, A., De Sisto, M., Nugues, L., Haider, T., Kočnik, N., Nagy, B., Delente, É., Renault, R., Bobenhausen, K., Hammerich, B., Mittmann, A., Palkó, G., Horváth, P., Navarro Colorado, B., Ruiz Fabo, P., Bermúdez Sabel, H., Korchagin, K., Plungian, V., Sitchinava, D. (2023). PoeTree. Poetry Treebanks in Czech, English, French, German, Hungarian, Italian, Portuguese, Russian, Slovenian, and Spanish (0.0.2) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.10008458

---

2. PoeTree Primary Journal Article

@article{poetree2024,
  author     = {Petr Plecháč and 
               Silvie Cinková and 
               Robert Kolár and 
               Artjoms Šeļa and 
               Mirella De Sisto and 
               Lara Nugues and 
               Thomas Haider and 
               Neža Kočnik},
  title      = {PoeTree: Poetry Treebanks in Czech, English, French, German, 
               Hungarian, Italian, Portuguese, Russian, Slovenian and Spanish},
  journal    = {Research Data Journal for the Humanities and Social Sciences},
  year       = {2024},
  publisher  = {Brill},
  address    = {Leiden, The Netherlands},
  doi        = {10.1163/24523666-bja10044},
  pages      = {1--17},
  url        = {https://brill.com/view/journals/rdj/aop/article-10.1163-24523666-bja10044/article-10.1163-24523666-bja10044.xml}
}

APA Citation:

Plecháč, P., Cinková, S., Kolár, R., Šeļa, A., De Sisto, M., Nugues, L., Haider, T., & Kočnik, N. (2024). PoeTree: Poetry Treebanks in Czech, English, French, German, Hungarian, Italian, Portuguese, Russian, Slovenian and Spanish. Research Data Journal for the Humanities and Social Sciences, 1–17. https://doi.org/10.1163/24523666-bja10044

---

3. Individual Language Corpora Citations

When using other subcorpora from the broader PoeTree collection, please refer to the corresponding standalone references:

• Czech (PoeTree.cs): Plecháč, P., & Kolár, R. (2015). The Corpus of Czech Verse. Studia Metrica et Poetica, 2(1), 107–118. https://doi.org/10.12697/smp.2015.2.1.05
• German (PoeTree.de):
  • Bobenhausen, K., & Hammerich, B. (2015). Métrique littéraire, métrique linguistique et métrique algorithmique de l'allemand mises en jeu dans le programme Metricalizer². Langages, (199), 67–87. https://doi.org/10.3917/lang.199.0067
  • Haider, T. (2021). Metrical Tagging in the Wild: Building and Annotating Poetry Corpora with Rhythmic Features. Proceedings of EACL 2021, 3715–3725. https://doi.org/10.18653/v1/2021.eacl-main.325
• English (PoeTree.en): Derived from Project Gutenberg ([no standalone paper reference; open access public domain]).
• Spanish (PoeTree.es):
  • Navarro-Colorado, B., Ribes Lafoz, M., & Sánchez, N. (2016). Metrical Annotation of a Large Corpus of Spanish Sonnets. Proceedings of LREC 2016, 4360–4364.
  • Ruiz Fabo, P., Bermúdez Sabel, H., Martínez Cantón, C., & González-Blanco, E. (2020). The Diachronic Spanish Sonnet Corpus. Digital Scholarship in the Humanities, 36(Suppl_1), i68–i80. https://doi.org/10.1093/llc/fqaa035
• French (PoeTree.fr): Delente, É., & Renault, R. (2021). Projet Anamètre : présentation, limites et avancées. In Plotting Poetry, On Mechanically-Enhanced Reading (pp. 73–92). Presses universitaires de Liège.
• Hungarian (PoeTree.hu): Horváth, P., et al. (2022). ELTE Poetry Corpus: A Machine Annotated Database of Canonical Hungarian Poetry. Proceedings of LREC 2022, 3471–3478. https://aclanthology.org/2022.lrec-1.372
• Italian (PoeTree.it): Quondam, A. Una storia lunga trenta anni. Biblioteca Italiana. http://www.bibliotecaitaliana.it/page/142028
• Norwegian (PoeTree.no): Kvinnsland, R., Dale, I. L., & Tungland, L. M. (2024). Rediscovering the 1890s: A Norwegian Poetry Corpus. Proceedings of CHR 2024, 1259–1271. https://ceur-ws.org/Vol-3834/#paper24
• Portuguese (PoeTree.pt): Mittmann, A., Pergher, P. H., & Luiz dos Santos, A. (2019). What Rhythmic Signature Says About Poetic Corpora. In Quantitative Approaches to Versification (pp. 153–172). ICL CAS.
• Russian (PoeTree.ru): Grishina, E., Korchagin, K., Plungian, V., & Sitchinava, D. (2009). Poeticheskii korpus v ramkah NKRIA: obschaia struktura i perspektivy ispolzovania. In Natsionalnii korpus russkogo iazyka: 2006–2008 (pp. 71–113). Nestor-Istoria.

---

PoeTree Project & Funding Information

• Main Portal: https://versologie.cz/poetree/
• Host Institution: Institute of Czech Literature of the Czech Academy of Sciences (ICL CAS) / Versologie Project
• Support & Funding: Supported by the Czech Science Foundation (Grant No. GA23-07727S)