This vignette provides a quick introduction to basic usage of the {tweetbotornot2} package. It introduces the two key functions for generating estimates from the built-in bot classifier.

## load package

Bot Classifier

The main use case for {tweetbotornot} is application of the built-in Twitter bot classifier. There are two key functions for getting bot-probability estimates for one or more Twitter accounts. Those functions are predict_bot() and predict_bot_score(). Both of these functions return estimates generated via same classifier, but predict_bot() returns a data.table with one row for each unique user and three columns containing the user_id, screen_name, and prob_bot (the model estimate). While predict_bot_score() returns a numeric vector of estimates matched to the input order of users.

## users
users <- c("netflix_bot", "PatrickMahomes", "PatrickMahomes", "netflix_bot")

## returns a data.table with 2 rows

## returns a numeric vector with 4 estimates

Of course, one thing missing from the above code are the actual Twitter data (or features) used to make the predictions. That’s because when screen names (or user IDs) are provided, {tweetbotornot2} does that work with the help of rtweet behind the scenes. But for that to work, users must obtain the proper authorization for accessing Twitter’s REST API. Fortunately, for most users this process should now be a breeze–e.g., running the code above may require one-click authorizing {rtweet}’s embedded rstats2twitter application but will otherwise just work. However, for users working in the cloud or trying to generate estimates for thousands or millions of uers, additional consideration may be warranted in terms of token-handling (see the "tokens" vignette) and rate limit-management (see the "bulk" vignette).

1 Unless you’ve already collected user timeline data, {tweetbotornot2} relies on {rtweet} to pull data from Twitter’s REST API.

2 The first time you make a request [from your computer] to Twitter’s API’s, a browser should pop-up, asking you to authorize the app. After that, the authorization token is stored and remembered behind the scenes (you will only have to reauthorize the app if your Twitter settings or token credentials get reset or if you use a different machine).