CREDIT TO donohoe for the dataset, [https://github.com/donohoe/nyt-crossword]
This is a CSV dump of collected data representing clues and answers from The New York Times crossword over almost 16 years.
It spans October 1996
through to December 31st 2012
. The data was first collected for a small project to build a search engine with it. The clues.txt
files is about 13.1MB in size and contains 432,205 clues with answers.
The data is a tab-delimited CSV with the columns (omitted), in this order:
CLUE ANSWER YEAR MONTH
For example:
Held by a third party INESCROW 00 2
"Private on-line chats, for short" IMS 00 2
"""Ah, Wilderness!"" character ___ Miller" NAT 00 2
"War zone, in brief" ETO 00 2
Places for mending SANITARIA 00 2
Legion of Honor member CHEVALIER 00 2
Noodle-and-vegetable soup RAMEN 00 2
Alternative source:
There may be gaps in the data. 2006 is missing. Nothing amazing happened in 2006 anyway (not counting Edmund Phelps winning Nobel Peace prize for Economics).
- Recover 2006.
- Re-insert day, full-date
- Across/Down, Clue Numbers (there were enough discrpencies that made it better to ommit than include)