Package 'textometry'

Title: Textual Data Analysis Package Used by the TXM Software
Description: Statistical exploration of textual corpora using several methods from French 'Textometrie' (new name of 'Lexicometrie') and French 'Data Analysis' schools. It includes methods for exploring irregularity of distribution of lexicon features across text sets or parts of texts (Specificity analysis); multi-dimensional exploration (Factorial analysis), etc. Those methods are used in the TXM software.
Authors: Sylvain Loiseau, Lise Vaudor, Matthieu Decorde, Serge Heiden
Maintainer: Matthieu Decorde <[email protected]>
License: GPL (>= 3)
Version: 0.1.6
Built: 2024-10-27 05:34:10 UTC
Source: https://github.com/cran/textometry

Help Index

Textual Data Analysis Package used by the TXM Software

Description

Statistical exploration of textual corpora using several methods from French 'Textometrie' (new name of 'Lexicometrie') and French 'Data Analysis' schools. It includes methods for exploring irregularity of distribution of lexicon features across text sets or parts of texts (Specificity analysis); multi-dimensional exploration (Factorial analysis), etc. Those methods are used in the TXM software.

Details

Package: textometry
Type: Package
Version: 0.1.3
Date: 2014-06-16
License: GPLv3
Depends: R (>= 1.5.0)

Index: 

specificities    Compute Lexical Specificity of subcorpus 
progression    Draw progression graphic

Author(s)

Sylvain Loiseau, Lise Vaudor, Matthieu Decorde, Lise Vaudor

Examples

data(robespierre);
specificities(robespierre);

adverbs frequency from 5 different domains of the BFM database

Description

A lexical table containing frequencies of adverbs from the BFM (Base de Francais m\'edi\'eval) database in 5 different domains (literary, historical, didactic, law, religious).

Usage

data(bfm)

Format

The format is: num [1:2, 1:5] 103000 1370887 23429 413441 15345 ... - attr(*, "dimnames")=List of 2 ..$ : chr [1:2] "ADV" "other" ..$ : chr [1:5] "literary" "history" "didactic" "juridical" ...

Details

The last line of the table gives the total frequency of all the other part of speech words in each of these domains.

Source

BFM: https://bfm.ens-lyon.fr

References

BFM - Base de Fran\,cais M\'edi\'eval [En ligne]. Lyon : ENS de Lyon, Laboratoire ICAR, 2012, https://bfm.ens-lyon.fr.

Draw progression graphic

Description

Draw the progression graphic of matches of CQL queries in a corpus

Usage

progression(positions, names, colors, styles, widths, corpusname, Xmin, T, 
  	doCumulative, structurepositions, strutnames, graphtitle, bande)

Arguments

positions

Vector containing corpus positions of CQL queries matches. A position is an integer from 0 (begining of corpus) to N (end of corpus)

names

String vector containing the CQL queries

colors

Vector containing the line color of each query

styles

Vector containing the line style of each query

widths

Vector containing the line width of each query

corpusname

String: corpus name

Xmin

Integer: corpus starting position of abscissa values

T

Integer: size of the corpus

doCumulative

Boolean: if true draw a cumulative graph, if false draw a density graph

structurepositions

optional Vector containing the structure positions of the corpus

strutnames

optional Vector containing the structures labels to display

graphtitle

String: graph title

bande

Float: density window size factor

Author(s)

Matthieu Decorde

5 words from Robespierre's discourses

Description

A lexical table containing frequencies of 5 words from 9 different public discourses of French politician Robespierre (between november 1793 and july 1794).

Usage

data(robespierre)

Format

The format is: num [1:6, 1:10] 464 45 35 30 6 ... - attr(*, "dimnames")=List of 2 ..$ : chr [1:6] "de" "peuple" "republique" "ennemi" ... ..$ : chr [1:10] "D1" "D2" "D3" "D4" ...

Details

The last line of the table gives the total frequency of all the other forms in each of these discourses.

Source

Lafon P. (1980) Sur la variabilit\'e de la fr\'e quence des formes dans un corpus, Mots, 1, pp. 127–165.

References

Lafon P. (1980) Sur la variabilit\'e de la fr\'e quence des formes dans un corpus, Mots, 1, pp. 127–165.

Examples

data(robespierre)

## See graphic in Lafon, 1980 - page 140

t <- colSums(robespierre)["D9"];     # size of the part
T <- sum(robespierre);               # size of the corpus
f <- rowSums(robespierre)["peuple"]; # total frequency of "peuple"
p <- dhyper(1:15, f, T-f, t)
title <- "Probability of each frequency of 'peuple' in the 'D9' discourse from 1 to 15"
plot(p, type="h", main=title, xlab="k", ylab="Prob(k)");

Calculate Lexical Specificity Score

Description

Calculate the specificity - or association or surprise - score of a word being present f times or more in a sub-corpus of t words given that it appears a total of F times in a whole corpus of T words.

Usage

specificities(lexicaltable, types=NULL, parts=NULL)

Arguments

lexicaltable

a complete lexical table, i.e. a numeric matrix where each line represents a word and each column a part of the corpus. Each cell gives the frequency of the given word in the corresponding part of the corpus.

types

list of rows (words) for which the specificity score must be calculated. If NULL, the specificity score is calculated for every row; If types is a character vector, it indicates the row names for which the specificity score is to be calculated (an error is thrown if lexicaltable has no row names); If types is an integer vector, it indicates the index of rows for which the specificity score is to be calculated.
parts

list of columns (parts) for which the specificity score must be calculated. If NULL, the specificity index is calculated for every part; If parts is a character vector, it indicates the column names for which the specificity score is to be calculated (an error is thrown if lexicaltable has no column names); If parts is an integer vector, it indicates the index of columns for which the specificity score is to be calculated.

Value

Returns a matrix of nrow(lexicaltable) * ncol(lexicaltable) (the number of rows and columns may be reduced using types or parts), each cell giving the specificity score.

Author(s)

Matthieu Decorde, Serge Heiden, Sylvain Loiseau, Lise Vaudor

References

Lafon P. (1980) Sur la variabilit\'e de la fr\'e quence des formes dans un corpus, Mots, 1, pp. 127–165. https://www.persee.fr/doc/mots_0243-6450_1980_num_1_1_1008

See Also

specificities.probabilities, specificities.lexicon

Examples

data(robespierre);
spe <- specificities(robespierre);
string <- paste("The word %s appears f=%d times in a sub-corpus of t=%d words,",
" given a total frequency of F=%d in the robespierre corpus made",
" of T=%d words. The corresponding specificity score is %f", sep="");
print(sprintf(string,
'peuple',
robespierre['peuple','D4'],
colSums(robespierre)['D4'],
rowSums(robespierre)['peuple'],
sum(robespierre),
spe['peuple', 'D4']));

Display specificities probability

Description

Display specificities probability distribution (call dhyper and specificities.probabilities.vector)

Usage

specificities.distribution.plot(x, F, t, T)

Arguments

x

observed number of A words
F

total number of A
t

size of part
T

size of corpus

Value

nothing

Author(s)

Matthieu Decorde, Serge Heiden

*OBSOLETE FUNCTION (see 'specificities.lexicon.new')* specificities association score with two frequency lists.

Description

Compute specificities association score between a lexicon and a sub-lexicon

Usage

specificities.lexicon(lexicon, sublexicon)

Arguments

lexicon

a frequency list (named vector)

sublexicon

a frequency list (named vector)

Value

specificities index as a named vector.

See Also

specificities for specificities score and references

specificities association score with two frequency list.

Description

Compute specificities association score between a lexicon and a sub-lexicon. A new version of the "specificities.lexicon" function

Usage

specificities.lexicon.new(lexicon, sublexicon)

Arguments

lexicon

a frequency list (named vector)

sublexicon

a frequency list (named vector)

Value

specificities index as a named vector.

See Also

specificities for specificities score and references

Calculate specificity probabilities

Description

Utility function computing specificity probabilities for the specificities function.

Usage

specificities.probabilities(lexicaltable, types = NULL, parts = NULL)

Arguments

lexicaltable

see specificities

types

see specificities

parts

see specificities

Value

Returns a matrix of signed specificity probabilities (between -1.0 and 1.0). By convention:

sign

The sign indicates if the observed frequency is lower (minus) or higher (plus) than the mode of the specificity model
.Machine$double.xmin limit

-10.0 and 10.0 values are used to hold the sign when the zero/.Machine$double.xmin boundary line has been crossed (the phyper function always returns 0.0)

See Also

see specificities.

Vector raw hypergeometric probabilities

Description

Calculate specificity probabilities on vector (call phyper and phyper_right)

Usage

specificities.probabilities.vector(v_f, v_F, T, t)

Arguments

v_f

vector of lexicon ferquencies
v_F

vector of corpus frequencies
T

corpus size
t

sub-corpus size

Value

Hypergeometric probabilities. See specificities.lexicon.

Author(s)

Matthieu Decorde, Serge Heiden