1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
|
\name{textual}
\alias{textual}
\title{Text mining}
\description{
Calculates the number of occurence of each words and a contingence table
}
\usage{
textual(tab, num.text, contingence.by=1:ncol(tab),
maj.in.min = TRUE, sep.word=NULL)
}
\arguments{
\item{tab}{a data frame with one textual variable}
\item{num.text}{indice of the textual variable}
\item{contingence.by}{a list with the indices of the variables for which a contingence
table is calculated by default a contingence table is calculated for all the variables
(except the textual one). A contingence table can also be calculated for couple of variables.
If \code{contingence.by} is equal to num.text, then the contingence table is calculated
for each row of the data table}
\item{maj.in.min}{boolean, if TRUE majuscule are transformed in minuscule}
\item{sep.word}{a string with all the characters which correspond to separator of words}
}
\value{
Returns a list including:
\item{cont.table}{the contingence table with in rows the categories of the categorical variables
(or the couple of categories), and in column the words,
and in each cell the number of occurence}
\item{nb.words}{a data.frame with all the words and for each word, the number of lists in which
it is present, and the number of occurence}
}
\author{Francois Husson \email{francois.husson@institut-agro.fr}}
\seealso{ \code{\link{CA}}, \code{\link{descfreq}}}
\examples{
data(poison.text)
res.text <- textual(poison.text, num.text = 3, contingence.by = 1)
descfreq(res.text$cont.table)
## Contingence table for the couple of variable sick-sex
res.text2 <- textual(poison.text, num.text = 3, contingence.by = list(c(1,2)))
descfreq(res.text2$cont.table)
## Contingence table for sex, sick and the couple of variable sick-sex
res.text2 <- textual(poison.text, num.text = 3, contingence.by = list(1,2,c(1,2)))
}
\keyword{multivariate}
|
