SemCor / README.md
MarkChen1214's picture
Update README.md
e4f47f9
metadata
dataset_info:
  features:
    - name: ID
      sequence: int64
    - name: Word
      sequence: string
    - name: Lemma
      sequence: string
    - name: POS
      sequence: string
    - name: Definition
      sequence: string
    - name: Lemma_sentence
      dtype: string
    - name: sentence
      dtype: string
    - name: Lemma_tfidf
      sequence: string
    - name: Lemma_tfidf_value
      sequence: float64
  splits:
    - name: train
      num_bytes: 24209901
      num_examples: 20138
  download_size: 8568417
  dataset_size: 24209901
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
license: mit
task_categories:
  - text-classification
language:
  - en
size_categories:
  - 1K<n<10K

Dataset Card for "SemCor – sense-tagged English corpus"

Description

This dataset is derived from the wsd_semcor dataset, originally hosted on Hugging Face. It has been preprocessed for tasks related to Word Sense Disambiguation (WSD) and WordNet integration.

Preprocessing

The original text data underwent the following preprocessing steps:

  • Text splitting into individual words (lemmas).
  • TF-IDF (Term Frequency-Inverse Document Frequency) analysis to understand the importance of words within the documents.

Structure

The dataset contains:

  • Lemmas: Words obtained from splitting the text data.
  • TF-IDF values: Quantitative measures of word importance within the documents.

Note

The number of elements in Lemma and Lemma_tfidf might not match. This is because Lemma is based on original dataset and might contain compound words, which might not be recognized by TF-IDF algorithm.

Intended Use

This dataset is intended for use in WSD and WordNet integration tasks. It provides foundational data for natural language processing (NLP) research and applications, specifically focusing on understanding word meanings and contextual usage.

Citation

Data sourced from wsd_semcor dataset on Hugging Face.

More Information needed