KDViz: Knowledge domain visualization
In andresfpc/KDViz: Knowledge Domain Visualization

Description Usage Arguments Value Examples

Knowledge domain visualization, to perform a SCA of each partitioned document-term matrix

1 2	KDViz(dtmGroup, graph = FALSE, ex = 1, ey = 2, ucal = 0, cex.row = 0.6, cex.col = 0.7)

`dtmGroup`	a document-term matrix
`graph`	a logical value, if TRUE a graph is displayed
`ex`	number identifying the factor to be used as horizontal axis (1 by default)
`ey`	number identifying the factor to be used as vertical axis (1 by default)
`ucal`	quality representation threshold (percentage) in the plane (0 by default)
`cex.row`	scale for row points and row labels (0.6 by default)
`cex.col`	scale for column points and column labels (0.7 by default)

the SCA from a document-term matrix group

## Not run: 
rm(list = ls())
library("KDViz")

risFile <- system.file("ScienceDirectRIS.ris", package = "KDViz") # Original data

myData <- ReadRIS(risFile, "bibDataRIS", saveRda = TRUE, saveCSV = FALSE) # RIS file to data object

bibData <- system.file("bibData.Rda", package = "KDViz")
load(bibData)


# Create a corpus from the bib data
corpus <- CorpusFromBibData(bibData = bibData, bibUnits = c("Keywords"),
   controlList = "", stopwords = "", wordsToRemove = "")

dtm <- DTMFromCorpus(corpus, row.names(bibData)) # Create a doc-term matrix from the corpus
dim(dtm)

# A first review of the raw corpus

bibUnits <- c("Keywords") # Selection of bibliometric units to analyze
controlList <- c("stripWhitespace", "removeNumbers") # List of tm process to perform
# Decide which stopwords are going to be used (a file or FALSE if they are not required)
stopwords <- FALSE
#stopwords <- system.file("stopwords_en.txt", package = "KDViz") # Optional
wordsToRemove <- c("nanotechnology") # List of custom words to remove

# Custom dictionary to replace some selected words
replaceWords <- system.file("keywordReplace.txt", package = "KDViz")

# Corpus from bibdata with and a control list to perform the entire tm process
corpus <- CorpusFromBibData(bibData = bibData, bibUnits = bibUnits,
   controlList = controlList, stopwords = stopwords,
   wordsToRemove = wordsToRemove, replaceWords = replaceWords)

termFreqTable <- TermFrequency(corpus) # See the frequency of terms in the corpus
head(termFreqTable, 98)

# Search for words containing the term in 'word' parameter
TermFreqByWord(termFreqTable = termFreqTable, word = "reduction")

# An optional function (contained yet in the previous process) to
#  replace other words after getting a corpus
#corpus <- ReplaceByList(corpus = corpus, wordsFile = replaceWords)

termFreqTable <- TermFrequency(corpus) # See the frequency of terms in the current corpus
head(termFreqTable, 100)

dtm <- DTMFromCorpus(corpus, row.names(bibData)) # Create a doc-term matrix from the corpus
dim(dtm)
rownames(dtm)

termFreq <- TermFrequency(dtm) # See the frequency of terms in the doc-term matrix
head(termFreq, 100)

mpaWords <- matriz.mpa.corpus(corpus, fmin = 5, cmin = 1) # mpa matrices from a corpus object
mpaWords$Palabras

# mpa method from the calculated objects in 'mpaWords'
classes <- mpa::mpa(mpaWords$Matriza, 10, mpaWords$Palabras)
classes

kdSummary <- KDSummary(matriz.mpa = mpaWords, mpa = classes) # a quick summary of the mpa process

mpa::plotmpa(3, mpaWords$Matriza, classes) # Plot the network of selected class

# Extract a partition of the original 'dtm' matrix depending on the class that you want
WordGroupDTM(dtm, wordClasses = kdSummary$wordClasses, class = 7, graph = TRUE)

group1 <- WordGroupDTM(dtm, wordClasses = kdSummary$wordClasses, class = 1, graph = TRUE)
group2 <- WordGroupDTM(dtm, wordClasses = kdSummary$wordClasses, class = 2, graph = TRUE)
group3 <- WordGroupDTM(dtm, wordClasses = kdSummary$wordClasses, class = 3, graph = TRUE)
group4 <- WordGroupDTM(dtm, wordClasses = kdSummary$wordClasses, class = 4, graph = TRUE)
group5 <- WordGroupDTM(dtm, wordClasses = kdSummary$wordClasses, class = 5, graph = TRUE)
group6 <- WordGroupDTM(dtm, wordClasses = kdSummary$wordClasses, class = 6, graph = TRUE)
group7 <- WordGroupDTM(dtm, wordClasses = kdSummary$wordClasses, class = 7, graph = TRUE)
group8 <- WordGroupDTM(dtm, wordClasses = kdSummary$wordClasses, class = 8, graph = TRUE)
group9 <- WordGroupDTM(dtm, wordClasses = kdSummary$wordClasses, class = 9, graph = TRUE)
group10 <- WordGroupDTM(dtm, wordClasses = kdSummary$wordClasses, class = 10, graph = TRUE)

plot(group1$coaGroup, ucal = 0, cex.col = 0.8, cex.row = 0.5)

LoadArticle(bibData, "A625") # Load the info of an article (it will open the URL by default)

## End(Not run)