metadata

license: mit
language:
  - en
task_categories:
  - text-classification
  - token-classification
  - other
task_ids:
  - named-entity-recognition
  - text-scoring
pretty_name: Dubliners (James Joyce)
description: >
  A dataset of James Joyce's collection of short stories "Dubliners," prepared
  for NLP tasks and computational analysis of literary texts. The dataset
  includes:

  - Text tokenized by sentences.

  - POS-tagged sentences using NLTK.

  - Results of analyzing the text with spaCy (POS-tagged, named entities,
  dependencies).


  This dataset was created as part of an NLP course at the Higher School of
  Economics (HSE). For more details, see the original repository:
  https://github.com/vifirsanova/compling.


  The dataset can be used for various NLP tasks, including:

  - Part-of-speech tagging.

  - Named entity recognition.

  - Dependency parsing.

  - Computational analysis of literary texts.


  It is particularly suited for researchers and students interested in
  computational linguistics and literary analysis.
size_categories:
  - 10K<n<100K
source_datasets:
  - original
dataset_info:
  features:
    - name: text
      dtype: string
      description: Raw text from "Dubliners," tokenized by sentences.
    - name: nltk_pos
      dtype: string
      description: Part-of-speech tags for each sentence, generated using NLTK.
    - name: spacy_pos
      dtype: string
      description: Part-of-speech tags for each sentence, generated using spaCy.
    - name: named_entities
      dtype: string
      description: Named entities identified in the text, generated using spaCy.
    - name: dependencies
      dtype: string
      description: Dependency parses for each sentence, generated using spaCy.
  splits:
    - name: train
      num_bytes: 5717280
      num_examples: 3949
  download_size: 5717280
  dataset_size: 5717280
tags:
  - literature
  - nlp
  - pos-tagging
  - named-entity-recognition
  - dependency-parsing
  - james-joyce
  - dubliners
  - computational-linguistics

Dataset Card for Dubliners (James Joyce)

Dataset Description
Dataset Structure
Usage
License
Citation

Dataset Description

Homepage: GitHub Repository
Repository: GitHub
Point of Contact: [[email protected]]
License: MIT

Dataset Summary

This dataset contains James Joyce's collection of short stories "Dubliners," prepared for NLP tasks and computational analysis. It includes:

Text tokenized by sentences.
POS-tagged sentences using NLTK.
Results of analyzing the text with spaCy (POS-tagged, named entities, dependencies).

Supported Tasks

Part-of-speech tagging
Named entity recognition
Dependency parsing
Computational analysis of literary texts

Dataset Structure

Data Fields

text: Raw text from "Dubliners," tokenized by sentences.
nltk_pos: Part-of-speech tags for each sentence, generated using NLTK.
spacy_pos: Part-of-speech tags for each sentence, generated using spaCy.
named_entities: Named entities identified in the text, generated using spaCy.
dependencies: Dependency parses for each sentence, generated using spaCy.

Data Splits

train: Contains the entire dataset.

Usage

This dataset is intended for use in NLP tasks such as part-of-speech tagging, named entity recognition, dependency parsing, and computational analysis of literary texts. It is particularly suited for researchers and students interested in computational linguistics and literary analysis.

License

This dataset is licensed under the MIT License.

Citation

If you use this dataset, please cite the original source:

@misc{dubliners-nlp-dataset,
  author = {doc_sportello},
  title = {Dubliners (James Joyce) NLP Dataset},
  year = {2025},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/docsportellochrys/nlp-learning}},
}

Datasets:

chryskylodon
/

dubliners