A word cloud is a collection of words. A word cloud is a simple yet powerful visual representation object for text processing. Text is more frequent and the more important it is, the bigger and bolder the word are shown.
Today I will try to make word cloud using titles of research articles. Dataset is "mantle xenolith" from PetDB.
Package
library(tm)
library(SnowballC)
library(wordcloud)
library(RColorBrewer)
library(tidyverse)
library(dplyr)Read the text file
data <- read.csv("petdb_mantle_xenolith_title.csv")Screening the data
Since there are too many data, I chose the articles published in LITHOS
data_new <- data %>% filter(Journal == "LITHOS")
text <- data_new[,2] #titles of articlesLoad the data as a corpus
docs <- Corpus(VectorSource(text))Inspect the content of the docment
#Inspect the content of the docment
inspect(docs)## <<SimpleCorpus>>
## Metadata: corpus specific: 1, document level (indexed): 0
## Content: documents: 177
##
## [1] A FORE-ARC SETTING OF THE GERF OPHIOLITE, EASTERN DESERT, EGYPT: EVIDENCE FROM MINERAL CHEMISTRY AND GEOCHEMISTRY OF ULTRAMAFITES
## [2] SERPENTINIZATION AND DEHYDRATION IN THE UPPER MANTLE BENEATH FUERTEVENTURA (EASTERN CANARY ISLANDS): EVIDENCE FROM MANTLE XENOLITHS
## [3] METASOMATISM IN LITHOSPHERIC MANTLE ROOTS: CONSTRAINTS FROM WHOLE-ROCK AND MINERAL CHEMICAL COMPOSITION OF DEFORMED PERIDOTITE XENOLITHS FROM KIMBERLITE PIPE UDACHNAYA
## [4] GEOCHEMISTRY OF ECLOGITE XENOLITHS FROM THE UDACHNAYA KIMBERLITE PIPE: SECTION OF ANCIENT OCEANIC CRUST SAMPLED
## [5] NATURE OF THE LITHOSPHERIC MANTLE BENEATH THE ARABIAN SHIELD AND GENESIS OF AL-SPINEL MICROPODS: EVIDENCE FROM THE MANTLE XENOLITHS OF HARRAT KISHB, WESTERN SAUDI ARABIA
## [6] PERSISTENCE OF FERTILE AND HYDROUS LITHOSPHERIC MANTLE BENEATH THE NORTHWESTERN ETHIOPIAN PLATEAU: EVIDENCE FROM MODAL, TRACE ELEMENT AND SR-ND-HF ISOTOPIC COMPOSITIONS OF AMPHIBOLE-BEARING MANTLE XENOLITHS
........ Transformation is performed using tm_map() function to replace, for example, special characters from the text
#Transformation is performed using tm_map() function to replace, for example, special characters from the text.
toSpace <- content_transformer(function (x , pattern ) gsub(pattern, " ", x))
docs <- tm_map(docs, toSpace, "/")
docs <- tm_map(docs, toSpace, "@")
docs <- tm_map(docs, toSpace, "\\|")Convert the text to lower case
# Convert the text to lower case
docs <- tm_map(docs, content_transformer(tolower))Remove numbers
# Remove numbers
docs <- tm_map(docs, removeNumbers)Remove english common stopwords
# Remove english common stopwords
docs <- tm_map(docs, removeWords, stopwords("english"))Remove punctuations
# Remove punctuations
docs <- tm_map(docs, removePunctuation)
# Remove your own stop word
# specify your stopwords as a character vector
#docs <- tm_map(docs, removeWords) Eliminate extra white spaces
# Eliminate extra white spaces
docs <- tm_map(docs, stripWhitespace)Text stemming
I will make word clouds with/without doing text stemming
# Text stemming
docs_new <- tm_map(docs, stemDocument)Build a term-document matrix
Document matrix is a table containing the frequency of the words. Column names are words and row names are documents.
dtm <- TermDocumentMatrix(docs)
m <- as.matrix(dtm)
v <- sort(rowSums(m),decreasing=TRUE)
d <- data.frame(word = names(v),freq=v)
head(d, 30)## word freq
## mantle mantle 133
## xenoliths xenoliths 104
## evidence evidence 44
## peridotite peridotite 42
## beneath beneath 38
## lithospheric lithospheric 31
## china china 27
## craton craton 26
## metasomatism metasomatism 21
## geochemistry geochemistry 20
## evolution evolution 19
## north north 19
## central central 18
## kimberlite kimberlite 16
## petrology petrology 16
## south south 16
## implications implications 16
## constraints constraints 15
## peridotites peridotites 15
## geochemical geochemical 15
## eastern eastern 14
## isotopic isotopic 14
## element element 13
## spinel spinel 13
## melt melt 12
## lithosphere lithosphere 12
## upper upper 11
## trace trace 11
## processes processes 11
## isotope isotope 11The result after text stemming
dtm <- TermDocumentMatrix(docs_new)
m <- as.matrix(dtm)
v <- sort(rowSums(m),decreasing=TRUE)
d_new <- data.frame(word = names(v),freq=v)
head(d_new, 30)## word freq
## mantl mantl 133
## xenolith xenolith 107
## peridotit peridotit 58
## evid evid 45
## lithospher lithospher 43
## beneath beneath 38
## metasomat metasomat 34
## isotop isotop 33
## craton craton 32
## china china 27
## kimberlit kimberlit 26
## melt melt 23
## petrolog petrolog 22
## geochemistri geochemistri 20
## evolut evolut 19
## north north 19
## composit composit 18
## central central 18
## element element 17
## south south 16
## implic implic 16
## constraint constraint 15
## geochem geochem 15
## eastern eastern 14
## spinel spinel 14
## miner miner 13
## upper upper 11
## trace trace 11
## process process 11
## northern northern 11Visualization 1
set.seed(1234)
wordcloud(words = d$word, freq = d$freq, min.freq = 1,
max.words=200, random.order=FALSE, rot.per=0.35,
colors=brewer.pal(8, "Dark2"))
Visualization 2 (color)
set.seed(1234)
wordcloud(words = d$word, freq = d$freq, min.freq = 1,
max.words=100, random.order=FALSE, rot.per=0.35,
colors=brewer.pal(8, "Spectral"))
Visualization 3-1-1 (color)
set.seed(1234)
wordcloud(words = d$word, freq = d$freq, min.freq = 1,
max.words=100, random.order=FALSE, rot.per=0.35,
colors=brewer.pal(8, "RdYlBu"))
Visualization 3-1-2 (word frequency)
d <- head(d, 50)
p <- ggplot(d, aes(x = freq, y=fct_reorder(word,freq), fill = ..x..)) +
geom_bar(stat = "identity")
p <- p + scale_fill_distiller(palette = "RdYlBu")
plot(p)
Visualization 3-2-1 (color)The result after text stemmingset.seed(1234)
wordcloud(words = d_new$word, freq = d$freq, min.freq = 1,
max.words=100, random.order=FALSE, rot.per=0.35,
colors=brewer.pal(8, "RdYlBu"))
Visualization 3-2-2 (word frequency)d_new <- head(d_new, 50)
p <- ggplot(d_new, aes(x = freq, y=fct_reorder(word,freq), fill = ..x..)) +
geom_bar(stat = "identity")
p <- p + scale_fill_distiller(palette = "RdYlBu")
plot(p)
References
http://www.sthda.com/english/wiki/text-mining-and-word-cloud-fundamentals-in-r-5-simple-steps-you-should-know http://rstudio-pubs-
static.s3.amazonaws.com/256588_57b585da6c054349825cba46685d8464.html https://stackoverflow.com/questions/36967573/stemming-words-using-tm-package-in-r-does-not-work-properly
){kind=link}
0 Comments