wordgraph (version 0.1.0)

add.centrality.variables.to.data.frame: Main function of the package. Creates the graphs and adds the centrality variables to the data frame.

Description

Creates the graph of all words contained in the wordvars, computes the centrality of each word and for each wordvars variable, a column containing the corresponding centrality score is added to the dataframe.

Supported centrality functions are "centrality_alpha", "centrality_authority", "centrality_betweenness", "centrality_closeness", "centrality_pagerank", "centrality_eigen".

After applying add.centrality.variables.to.data.frame, subsequent quantitative analysis can follow where each word is replaced by its correspondent centrality score.

Usage

add.centrality.variables.to.data.frame(
  wordvars,
  centralityfunctionstr,
  data.df,
  iscircled = FALSE,
  verbose = FALSE)

Arguments

wordvars

The vector containing the names of the variables containing the free associated words (from 1_st to n_th)

centralityfunctionstr

The type of the centrality function to use (single string or vector of strings)

data.df

The data frame where wordvars belong.

iscircled

Set to TRUE if the graph is circled (that is, last word is related to the first). Default is FALSE.

verbose

Set to TRUE if warning messages are desired. Default is FALSE.

Value

The initial data frame with additional columns, one for each variable in wordvars vector, having the correspondiding centrality of the contained words.

Examples

# NOT RUN {
data(freeassociationdata)

# It is a time consuming function...

# }
# NOT RUN {
wordvars = c("diet1stword", "diet2ndword", "diet3rdword")
df.with.centralities1 = add.centrality.variables.to.data.frame(wordvars,
"centrality_eigen", freeassociationdata)

# Variables with the centralities of the words in wordvars was added
# in the end of df.with.centralities1.

# Histogram of eigen centrality index
hist(df.with.centralities1$diet1stword_centrality_eigen)

# One may ask for two centralities...
df.with.centralities2 = add.centrality.variables.to.data.frame(wordvars,
c("centrality_authority", "centrality_betweenness"), freeassociationdata)

# Pearson correlation between authority and betweeness centrality indexes for the 1st word
# concerning diet.

cor(df.with.centralities2[c("diet1stword_centrality_authority",
"diet1stword_centrality_betweenness")], use="complete.obs", method = "pearson")

# ...or even all available centralities.
# Warning: depending on the number of variables and the size of the data frame, it may be a
# time consuming procedure...

df.with.centralities2 = add.centrality.variables.to.data.frame(wordvars,
"all", freeassociationdata)

# }