Dataset of user and tweet ids of followers of @nytimes.

## Data Format

There are three files containing the ids for different types of Twitter entities.
All files are [xz compressed archives](https://en.wikipedia.org/wiki/XZ_Utils) and
can be unpacked with, e.g., `unxz user_ids.dat.xz`. Each file contains one id per
line, which are sorted ascendingly to improve compressibility.

*  `user_ids.dat.xz`
*  `tweet_ids.dat.xz`
*  `retweet_ids.dat.xz`

The ids of `user_ids.dat.xz` correspond to user_ids of Twitter.
The ids of `tweet_ids.dat.xz` and `retweet_ids.dat.xz` correspond to tweet_ids of Twitter,
but for convenience we split categorized them into retweets and regular tweets.

Further there are two files mapping tweet ids to nyt ids, if they contain a link to an
article of the NYT.

*  `tweet_nyt_ids.dat.xz`
*  `tweet_nyturl_ids.dat.xz`

The file `tweet_nyt_ids.dat.xz` contains id from tweets tweeted by the @nytimes account and in the second column
the id of the linked article (or `None` if the tweet did not link to an article).
The file `tweet_nyturl_ids.dat.xz` contains tweet ids of tweets from @nytimes followers which
contain a link to an article of the NYT and in the second column the id of the article.