by twitter

R package twitteR (by Jeff Gentry)

The simplest and quickest way to get data from twitter is by using the

functions of the R package twitteR (check twitteR vignette for more info)

Some (but not all) of the things you can get with twitteR

public tweets

trending topics

tweets containing a given hashtag

tweets containing a given word / term

tweets from a given user

Let's see different examples

Start: First we need to load the package twitteR

(remember to install it first!)

# load package twitteR

library(twitteR)

Case 1: Getting recent public tweets with publicTimeline

(limited to 20 tweets by default)

# get recent public tweets

public_tweets = publicTimeline()

> head(public_tweets, 3)

[[1]]

[1] "Na_rinhaa: tava dormindo ate agora man"

[[2]]

[1] "thesillykid: Photo: http://t.co/E0FGXQKJ"

[[3]]

[1] "LilBuffy2Bhad: Because jr its the only way I can get too the weed. Y u ask so many questions about it.?"

Case 2: Getting trending topics with the function getTrends

(there's a limit of past days to retrieve information from; roughly 10-14 days)

# get trending topics of the day

# (returns top 20 trending topics per hour for given data)

today_trends = getTrends(period = "daily", date=Sys.Date())

> # 5 trends from last hour

> today_trends[1:5]

[[1]]

[1] "La #MarchaYoSoy132"

[[2]]

[1] "#ThingsThatMakeMeHulkMad"

[[3]]

[1] "#Sopio"

[[4]]

[1] "#LawsonAdamFollows6"

[[5]]

[1] "Of Mice and Men"

Case 3: Getting trending topics from yesterday

# get trending topics from yesterday

yest_trends = getTrends(period = "daily", date=Sys.Date() - 1)

> # 5 trends from 24 hours ago

> yest_trends[1:5]

[[1]]

[1] "#AlwaysShadyNeverBieber"

[[2]]

[1] "#LawsonAdamFollows7"

[[3]]

[1] "Zux"

[[4]]

[1] "#CongratsHarry"

[[5]]

[1] "Zayn and Lux"

Case 4: Getting trends of the week

# get trending topics of the week

# (returns top 30 trending topics for each day)

week_trends = getTrends(period = "weekly")

> week_trends[1:5]

[[1]]

[1] "La #MarchaYoSoy132"

[[2]]

[1] "51 Million Facebook Monsters"

[[3]]

[1] "Chelsea <3"

[[4]]

[1] "Goodmorning <3"

[[5]]

[1] "#HowToMakePeopleMad"

Case 5: Getting tweets containing a given hashtag with searchTwitter

You can also search a maximum of 'n' tweets in a given language (given by an ISO 639-1 code)

For instance, get tweets with "#datamining"

# Search tweets with '#datamining'

dmhash_tweets = searchTwitter("#datamining")

# search a maximum of 20 tweets in french

dm20fr_tweets = searchTwitter("#datamining", lang="fr", n=20)

# search a maximum of 10 tweets in german

dm10de_tweets = searchTwitter("#datamining", lang="de", n=10)

> dm20fr_tweets

[[1]]

[1] "Alice_Begue: Le #Datamining pour les nuls http://t.co/jvYJKUMv"

[[2]]

[1] "marionlapeyroux: Le #Time2market, c'est cette semaine, rdv &amp; #conférences avec les spécialistes du #marketing #digital, #datamining.. http://t.co/R3os0Yro"

[[3]]

[1] "zeggar: RT @DZSSUG: Un Club de rugby utilise #BusinessIntelligence et #DataMining pour prévenir les blessures de ses joueurs. http://t.co/tBsksJWl"

> dm10de_tweets

[[1]]

[1] "WJastrinsky: Welche Trendfarben tragen die Menschen - in diesem Moment - in Paris? Color Forecast von pimkie. #datamining http://t.co/6vxTlqjx"

Case 6: Getting tweets containing a given word/term with searchTwitter

For instance, get tweets with "data mining"

# search tweets containing "data mining"

dm1_tweets = searchTwitter("data mining")

# search tweets between two dates (you will have to change the dates!)

dm2_tweets = searchTwitter("data mining", since='2012-05-12', until='2012-05-17')

# search tweets around a given radius of 5 miles of latitude/longitude

dm3_tweets = searchTwitter("data mining", geocode="37.857253,-122.270558,5mi")

Case 7: Get tweets from a given user with userTimeline

For instance, look for @weather's tweets

# tweets from @weather

wea_tweets = userTimeline("weather")

> wea_tweets[1:4]

[[1]]

[1] "weather: RT @science: Dramatic trailer for PBS Dust Bowl documentary http://t.co/i3wuASV9"

[[2]]

[1] "weather: Video: Twin waterspouts filmed near Grand Isle, LA. http://t.co/w99JkPUO"

[[3]]

[1] "weather: Prince Charles reads the Scottish weather forecast http://t.co/eUh8UdVo"

[[4]]

[1] "weather: Tornado damages 890 homes in Japan, kills teenager http://t.co/0TDZNYPs (via @newsalert)"