stopwords-en

Overview

The stopword-en dataset contains a stopword list of frequently used in the English language. These words do not carry significant meaning and are often removed from text data during preprocessing and training in shallower models on a text classification task.

Dataset Details

- Dataset Name: stopwords-en
- Total Size: 220 demonstrations

Contents

The dataset consists of one column with strings like all the letters of the Roman alphabet, numbers from 1 to 10, and words frequently used in the English language, such as "day", "days", "know", "went", "like", etc.

How to use

from sklearn.feature_extraction.text import TfidfVectorizer

# Download the English stopword list.
stopwords = load_dataset('AiresPucrs/stopwords-en', split='train')['stopwords']

# Create a vectorization object via `TfidfVectorizer`
vectorizer = TfidfVectorizer(min_df=10,
                             max_features=100000,
                             analyzer='word',
                             ngram_range=(1, 2),
                             stop_words=stopwords, # Our list of stopwords.
                             lowercase=True)

# Fit the TfidfVectorizer to our dataset.
vectorizer.fit(dataset['text'])

License

This dataset is licensed under the Apache License, version 2.0.