In livioivil/TextWiller: Collection of functions for text mining, specially devoted to the italian language

TextWiller

Collection of text mining utilities, specially devoted to the italian language.

---

Set up

To install this github version type (in R):

#if devtools is not installed yet: 
# install.packages("devtools")

#if you wanto to download the last stable version, use the following code
library(devtools)
install_github("livioivil/TextWiller@TextWiller_JOSS")

#if you want to download the latest release (unstable) use the following code
library(devtools)
install_github("livioivil/TextWiller",build_vignettes = TRUE)

---

Some examples

library(TextWiller)

### normalize texts
normalizzaTesti(c('ciao bella!','www.associazionerospo.org','noooo, che grandeeeeee!!!!!','mitticooo', 'mai possibile?!?!'))

# get the sentiment of a document
sentiment(c("ciao bella!","farabutto!","fofi sei figo!"))


# Classify users' gender by (italian) names
classificaUtenti(c('livio','alessandra','andrea'))
# and classify location
data(vocabolarioLuoghi)
classificaUtenti(c('Bosa','Pordenone, Italy','Milan'),vocabolarioLuoghi)


# find re-tweet (RT) by evaluation of texts similarity (and replace texts so that they become equals):
data(TWsperimentazioneanimale)
RTHound(TWsperimentazioneanimale[1:10,"text"], S = 3, L = 1, 
                 hclust.dist = 100, hclust.method = "complete",
                 showTopN=3)


#extract short urls and get the long ones
## Not run: urls=urlExtract("Influenza Vaccination | ONS - Oncology Nursing Society http://t.co/924sRKGBU9 See All http://t.co/dbtPJRMl00")

#extract users:
patternExtract(c("@luca @paolo: buon giorno!", "@matteo: a te!"), pattern="@\\w+")

livioivil/TextWiller documentation built on Nov. 30, 2020, 3:17 a.m.