<!-- README.md is generated from README.Rmd. Please edit that file -->

tweetbotornot2 <img src="man/figures/logo.png" width="160px" align="right" />

<!-- badges: start -->

<!-- badges: end -->

{tweetbotornot2} provides an out-of-the-box classifier for detecting Twitter bots that is easy to use, interpretable, scalable, and performant. It also provides a convenient interface for accessing the botometer API.

Installation

<!-- Install the released version of tweetbotornot2 from [CRAN](https://CRAN.R-project.org) with: --> <!-- ``` r --> <!-- ## install from CRAN --> <!-- install.packages("tweetbotornot2") --> <!-- ``` -->

Install the development version of {tweetbotornot2} from Github with:

## install {remotes} if not already
if (!"remotes" %in% installed.packages()) {
  install.packages("remotes")
}

## install from github
remotes::install_github("mkearney/tweetbotornot2")

Predict

Use predict_bot() to run the built-in bot classifier

Provide a vector or data frame of Twitter handles and predict_bot() will return the estimated probability of each account being a bot.

## vector of screen names
screen_names <- c(
  "American__Voter", ## (these ones should be bots)
  "MagicRealismBot",
  "netflix_bot",
  "mitchhedbot",
  "rstats4ds",
  "thinkpiecebot",
  "tidyversetweets",
  "newstarsbot",
  "CRANberriesFeed",
  "AOC",             ## (these ones should NOT be bots)
  "realDonaldTrump",
  "NateSilver538",
  "ChadPergram",
  "kumailn",
  "mindykaling",
  "hspter",
  "rdpeng",
  "kearneymw",
  "dfreelon",
  "AmeliaMN",
  "winston_chang"
)

## data frame with screen names **must be named 'screen_name'**
screen_names_df <- data.frame(screen_name = screen_names)

## vector -> bot estimates
predict_bot(screen_names)
#>                 user_id     screen_name   prob_bot
#>  1:  829792389925597184 American__Voter 0.99923730
#>  2:          3701125272 MagicRealismBot 0.99886143
#>  3:          1203840834     netflix_bot 0.85550964
#>  4:           214244836     mitchhedbot 0.99847370
#>  5: 1075011651366199297       rstats4ds 0.99878043
#>  6:          3325527710   thinkpiecebot 0.99953938
#>  7:  935569091678691328 tidyversetweets 0.99963319
#>  8:  780707721209188352     newstarsbot 0.99973100
#>  9:           233585808 CRANberriesFeed 0.99852484
#> 10:           138203134             AOC 0.00082178
#> 11:            25073877 realDonaldTrump 0.00126745
#> 12:            16017475   NateSilver538 0.00203745
#> 13:            16187637     ChadPergram 0.00385066
#> 14:            28406270         kumailn 0.00056573
#> 15:            23544596     mindykaling 0.00087570
#> 16:            24228154          hspter 0.00045269
#> 17:             9308212          rdpeng 0.00398646
#> 18:          2973406683       kearneymw 0.01408189
#> 19:            93476253        dfreelon 0.00055131
#> 20:            19520842        AmeliaMN 0.00769005
#> 21:          1098742782   winston_chang 0.00111468
#>                 user_id     screen_name   prob_bot

## data.frame -> bot estimates
#predict_bot(screen_names_df)

This also works on Twitter user IDs.

## vector of user IDs (strings of numbers, ranging from 2-19 digits)
user_ids <- rtweet::lookup_users(screen_names)[["user_id"]]

## data frame with user IDs **must be named 'user_id'**
user_ids_df <- data.frame(user_id = users)

## vector -> bot estimates
predict_bot(user_ids)

## data.frame -> bot estimates
predict_bot(user_ids_df)

The input given to predict_bot() can also be Twitter data returned by {rtweet}, i.e., rtweet::get_timelines()¹.

## timeline data returned by {rtweet}
twtdat <- rtweet::get_timelines(screen_names, n = 200, check = FALSE)

## generate predictions from twitter data frame
predict_bot(twtdat)
#>                 user_id     screen_name   prob_bot
#>  1:  829792389925597184 American__Voter 0.99923730
#>  2:          3701125272 MagicRealismBot 0.99886143
#>  3:          1203840834     netflix_bot 0.85550964
#>  4:           214244836     mitchhedbot 0.99847370
#>  5: 1075011651366199297       rstats4ds 0.99878043
#>  6:          3325527710   thinkpiecebot 0.99953938
#>  7:  935569091678691328 tidyversetweets 0.99963319
#>  8:  780707721209188352     newstarsbot 0.99973100
#>  9:           233585808 CRANberriesFeed 0.99852484
#> 10:           138203134             AOC 0.00082178
#> 11:            25073877 realDonaldTrump 0.00126745
#> 12:            16017475   NateSilver538 0.00203745
#> 13:            16187637     ChadPergram 0.00385066
#> 14:            28406270         kumailn 0.00056573
#> 15:            23544596     mindykaling 0.00087570
#> 16:            24228154          hspter 0.00045269
#> 17:             9308212          rdpeng 0.00398646
#> 18:          2973406683       kearneymw 0.01408189
#> 19:            93476253        dfreelon 0.00055131
#> 20:            19520842        AmeliaMN 0.00769005
#> 21:          1098742782   winston_chang 0.00111468
#>                 user_id     screen_name   prob_bot

Explain

Use explain_bot() to see the contributions made by each feature

View prediction contributions from top five features (for each user) in the model

## view top feature contributions in prediction for each user
explain_bot(twtdat)[
  order(screen_name, 
  -abs(value)), ][
    feature %in% feature[1:5],
    .SD, on = "feature" ][1:50, -1]
#>         screen_name   prob_bot   feature     value                feature_description
#>  1:             AOC 0.00082178 twt_srctw -4.074586 Tweet source of Twitter (official)
#>  2:             AOC 0.00082178 twt_srcna -0.788900            Tweet source of unknown
#>  3:             AOC 0.00082178 usr_fllws -0.539794                     User followers
#>  4:             AOC 0.00082178 twt_rtwts -0.453744                 Tweet via retweets
#>  5:             AOC 0.00082178 twt_quots -0.276252                   Tweet via quotes
#>  6:        AmeliaMN 0.00769005 twt_srctw -2.392487 Tweet source of Twitter (official)
#>  7:        AmeliaMN 0.00769005 twt_srcna -0.716127            Tweet source of unknown
#>  8:        AmeliaMN 0.00769005 twt_rtwts -0.461190                 Tweet via retweets
#>  9:        AmeliaMN 0.00769005 twt_quots -0.308175                   Tweet via quotes
#> 10:        AmeliaMN 0.00769005 usr_fllws  0.050839                     User followers
#> 11: American__Voter 0.99923730 twt_srctw  2.053514 Tweet source of Twitter (official)
#> 12: American__Voter 0.99923730 twt_srcna  1.149764            Tweet source of unknown
#> 13: American__Voter 0.99923730 twt_rtwts  0.357076                 Tweet via retweets
#> 14: American__Voter 0.99923730 usr_fllws  0.113606                     User followers
#> 15: American__Voter 0.99923730 twt_quots  0.020683                   Tweet via quotes
#> 16: CRANberriesFeed 0.99852484 twt_srctw  2.343053 Tweet source of Twitter (official)
#> 17: CRANberriesFeed 0.99852484 twt_srcna  1.026885            Tweet source of unknown
#> 18: CRANberriesFeed 0.99852484 twt_rtwts  0.340709                 Tweet via retweets
#> 19: CRANberriesFeed 0.99852484 usr_fllws  0.081496                     User followers
#> 20: CRANberriesFeed 0.99852484 twt_quots  0.009263                   Tweet via quotes
#> 21:     ChadPergram 0.00385066 twt_srctw -4.741660 Tweet source of Twitter (official)
#> 22:     ChadPergram 0.00385066 twt_srcna -0.573186            Tweet source of unknown
#> 23:     ChadPergram 0.00385066 twt_rtwts  0.470594                 Tweet via retweets
#> 24:     ChadPergram 0.00385066 usr_fllws -0.271190                     User followers
#> 25:     ChadPergram 0.00385066 twt_quots  0.016482                   Tweet via quotes
#> 26: MagicRealismBot 0.99886143 twt_srctw  2.114994 Tweet source of Twitter (official)
#> 27: MagicRealismBot 0.99886143 twt_srcna  1.112244            Tweet source of unknown
#> 28: MagicRealismBot 0.99886143 usr_fllws -0.596811                     User followers
#> 29: MagicRealismBot 0.99886143 twt_rtwts  0.321603                 Tweet via retweets
#> 30: MagicRealismBot 0.99886143 twt_quots