Data Collection

The data is parsed using ceshine/wickedQuotes, which is a fork of heyseth / wickedQuotes by Seth Miller.

I did some refactoring to the original code and convert the JSON dump into two CSV files (one containing source information, while the other containing the quotes).

Note that Seth Miller also published a Wikiquote dataset here on Kaggle, but it is in JSON format.

The 2020-01-20 Wikiquote dump is used to generate this dataset.

Additional File information

We provide two set of quotes:

• The quotes-100-en-* files contain quotes that are shorter than 100 characters.
• The quotes-500-en-* files contain quotes that are shorter than 500 characters.