4.5. MFACT Plane
Wine and Word Configuration on the First Plane Figure 2 shows the configuration of the eight wines on the first MFACT plane.
Global representation of the documents
plot.MFA(res.mfact.23,choix="ind",col.hab=rep("grey30",8),
title="Global representation of the wines-documents",cex=1,graph.type="classic")
Another global representation of the documents
dfp <- as.data.frame(res.mfact.23$ind$coord)
ax1 <-1 ; ax2 <-2
labx <- paste0("Dim 1 (", round(res.mfact.23$eig[ax1,2],1),"%)" )
laby <- paste0("Dim 2 (", round(res.mfact.23$eig[ax2,2],1),"%)" )
Figure2 <- ggplot(dfp, aes(x=Dim.1, y=Dim.2, label = rownames(dfp)))+
theme_light() + xlab(labx)+ ylab(laby) + coord_fixed()+
geom_hline(yintercept=0, linetype="dashed", color = "grey")+
geom_vline(xintercept=0, linetype="dashed", color = "grey")+
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())+
ggplot2::annotate("rect", xmin = c(-6), xmax = c(8),
ymin = -2.1, ymax = 2.2, alpha = .2, fill = c("green"),
color="black", linewidth=1) +
ggplot2::annotate("rect", xmin = c(-3), xmax = c(2),
ymin = -6.3, ymax = 5.8, alpha = .2, fill = c("brown"),color="black",
linewidth=1, lty="dashed")+
geom_text(size=5, fontface = "bold")+
theme(axis.title.x= element_text(size=15, face="bold"))+
theme(axis.title.y= element_text(size=15, face="bold"))+
theme(plot.margin = grid::unit(c(t=0,r= 2,b=5, l=2), "mm"))+
theme(axis.title.x = element_text(vjust = -2, hjust=0.5)) +
theme(panel.border = element_rect(colour = "black", fill=NA, linewidth=1))+
theme( axis.text.x = element_text(size=rel(1.6)))+
theme( axis.text.y = element_text(size=rel(1.6)))+
ggplot2::annotate("text", x=6.2,y=5.2,label = "Grenache wines", color ="black", size=5)+
ggplot2::annotate("text", x=6.2,y=4.6,label = "Carignan wines", color ="black", size=5)+
ggplot2::annotate("rect", xmin = c(3.5), xmax = c(3.9),
ymin = 4.9, ymax = 5.5, alpha = .2, fill = c("brown"),
color="black", linewidth=1, lty="dashed")+
ggplot2::annotate("rect", xmin = c(3.5), xmax = c(3.9),
ymin = 4.2, ymax = 4.8, alpha = .2, fill = c("green"), color="black", linewidth=1)+
ggtitle("Figure 2. Global representation of the wines documents")
Figure2
Characteristic words of the three poles
There are three clusters or poles in the first MFA plane:
- Cluster 1: (pole1) “PC06” “EC05” # Right of first factorial plane (first and second quadrants)
- Cluster 2: (pole2) “PG05” “PG06” “PC05” “EC06” # (Second quadrant)
- Cluster 3: (pole3) “EG05” “EG06” # Lower part of Dim 2 (third quadrant)
In order to position the characteristic words on the first factorial plane, the LexChar function of the Xplortext package will be used. Given the large number of words, only those with a minimum frequency of 3 for the French judges and 2 for the Catalan judges will be considered. A probability of 0.1 is established for this selection in order to have a reduced number of words and thus facilitate interpretation.
We must to remove FP2 judge (column 24) saving the results to new.base data frame.
The rows (wines) are: “PG05” “PG06” “EG05” “EG06” “PC05” “PC06” “EC05” “EC06”.
We have to add the cluster number (pole) to each of the wines.
new.base[,27]<-c(“Cl2”,“Cl2”,“Cl3”,“Cl3”,“Cl2”,“Cl1”,“Cl1”,“Cl2”)
colnames(base)
[1] "CE1" "CE2" "CE3" "CE4" "FE5" "FE6"
[7] "CE7" "CE8" "CE9" "CE10" "CE11" "FE12"
[13] "FP1" "FP3" "FP4" "FP5" "FP6" "FP7"
[19] "FP8" "FP9" "FP10" "FP11" "FP12" "FP2"
[25] "FrScore" "CatScore"new.base <- base[,-24]
colnames(new.base)
[1] "CE1" "CE2" "CE3" "CE4" "FE5" "FE6"
[7] "CE7" "CE8" "CE9" "CE10" "CE11" "FE12"
[13] "FP1" "FP3" "FP4" "FP5" "FP6" "FP7"
[19] "FP8" "FP9" "FP10" "FP11" "FP12" "FrScore"
[25] "CatScore"cat(dim(base))
8 26new.base$pole<-c("Cl2","Cl2","Cl3","Cl3","Cl2","Cl1","Cl1","Cl2")
colnames(new.base)
[1] "CE1" "CE2" "CE3" "CE4" "FE5" "FE6"
[7] "CE7" "CE8" "CE9" "CE10" "CE11" "FE12"
[13] "FP1" "FP3" "FP4" "FP5" "FP6" "FP7"
[19] "FP8" "FP9" "FP10" "FP11" "FP12" "FrScore"
[25] "CatScore" "pole" To take the positions of 14 French judges:
posit.Fr <- c(5,6,12:23)
names(new.base[, posit.Fr])
[1] "FE5" "FE6" "FE12" "FP1" "FP3" "FP4" "FP5" "FP6" "FP7" "FP8"
[11] "FP9" "FP10" "FP11" "FP12"To take the positions of 9 Catalan judges:
posit.Cat <- c(1:4, 7:11)
names(new.base[, posit.Cat])
"CE1" "CE2" "CE3" "CE4" "CE7" "CE8" "CE9" "CE10" "CE11"To obtain the French characteristic words for each cluster:
res.TD.Fr.Agg<-TextData(new.base,var.text=c(posit.Fr), stop.word.user=str.Fr.stopworduser,var.agg="pole",Fmin=3)
LC.Fig3.Fr <-LexChar(res.TD.Fr.Agg,proba=0.1)
LC.Fig3.Fr$CharWord
$Cl1
Intern % glob % Intern freq Glob freq p.value v.test
bois 4.4585987 1.2519562 7 8 0.0003207468 3.597941
boisé 10.1910828 4.6948357 16 30 0.0006021027 3.430665
neuf 2.5477707 0.6259781 4 4 0.0035396358 2.916517
dominé 1.9108280 0.4694836 3 3 0.0146182364 2.441701
charpenté 2.5477707 0.7824726 4 5 0.0151396078 2.429022
vanillé 3.8216561 1.7214397 6 11 0.0368077948 2.087890
finale 1.9108280 0.6259781 3 4 0.0513936738 1.948178
nonboisé 0.0000000 1.4084507 0 9 0.0776001980 -1.764783
animal 0.0000000 1.4084507 0 9 0.0776001980 -1.764783
fruit 1.2738854 4.2253521 2 27 0.0253952179 -2.235337
fruité 0.6369427 3.7558685 1 24 0.0101671008 -2.570094
$Cl2
Intern % glob % Intern freq Glob freq p.value v.test
fruit 6.508876 4.2253521 22 27 2.011013e-03 3.088601
fruité 5.325444 3.7558685 18 24 2.740705e-02 2.205671
souple 2.071006 1.2519562 7 8 5.463815e-02 1.921743
léger 2.071006 1.2519562 7 8 5.463815e-02 1.921743
rond 1.183432 0.6259781 4 4 7.762696e-02 1.764624
noir 1.183432 0.6259781 4 4 7.762696e-02 1.764624
mûr 2.958580 2.0344288 10 13 8.608338e-02 1.716430
fraîcheur 1.775148 1.0954617 6 7 9.388001e-02 1.675276
évent 1.775148 1.0954617 6 7 9.388001e-02 1.675276
tannique 1.183432 2.1909233 4 14 7.342496e-02 -1.790180
toasté 0.000000 0.6259781 0 4 4.871437e-02 -1.971083
neuf 0.000000 0.6259781 0 4 4.871437e-02 -1.971083
finale 0.000000 0.6259781 0 4 4.871437e-02 -1.971083
bois 0.295858 1.2519562 1 8 2.582280e-02 -2.228866
vanillé 0.591716 1.7214397 2 11 2.325646e-02 -2.269195
charpenté 0.000000 0.7824726 0 5 2.278452e-02 -2.277029
boisé 1.183432 4.6948357 4 30 5.537868e-06 -4.543306
$Cl3
Intern % glob % Intern freq Glob freq p.value v.test
toasté 2.083333 0.6259781 3 4 0.04004621 2.053272
asséchant 2.777778 1.0954617 4 7 0.05678802 1.904939
épice 0.000000 1.4084507 0 9 0.09879800 -1.650709
très 0.000000 2.0344288 0 13 0.03488995 -2.109633
attr(,"class")
[1] "descfreq" "list " To obtain the Catalan characteristic words for each cluster:
res.TD.Cat.Agg<-TextData(new.base,var.text=c(posit.Cat), stop.word.user=str.Cat.stopworduser,var.agg="pole",Fmin=2)
LC.Fig3.Cat <-LexChar(res.TD.Cat.Agg,proba=0.1)
LC.Fig3.Cat$CharWord
$Cl1
Intern % glob % Intern freq Glob freq p.value v.test
fusta 8.333333 2.7863777 7 9 0.001595904 3.156654
alt 4.761905 1.2383901 4 4 0.004334538 2.852730
cafè 3.571429 0.9287926 3 3 0.017124102 2.384032
especiat 4.761905 1.5479876 4 5 0.018411597 2.357235
torrat 3.571429 1.2383901 3 4 0.059827331 1.882064
claudolor 3.571429 1.2383901 3 4 0.059827331 1.882064
sutja 2.380952 0.6191950 2 2 0.067034594 1.831442
marcat 2.380952 0.6191950 2 2 0.067034594 1.831442
bota 2.380952 0.6191950 2 2 0.067034594 1.831442
regalèssia 4.761905 2.1671827 4 7 0.092345322 1.683154
madur 0.000000 2.4767802 0 8 0.087113463 -1.710825
fruita 0.000000 3.7151703 0 12 0.025018213 -2.241121
confitura 0.000000 4.0247678 0 13 0.018260882 -2.360285
$Cl2
Intern % glob % Intern freq Glob freq p.value v.test
fruita 6.0240964 3.715170 10 12 0.02588888 2.227875
fusta 1.2048193 2.786378 2 9 0.08878735 -1.701829
regalèssia 0.6024096 2.167183 1 7 0.05793784 -1.896168
astringent 0.6024096 2.167183 1 7 0.05793784 -1.896168
torrat 0.0000000 1.238390 0 4 0.05472216 -1.921076
taní 0.0000000 1.238390 0 4 0.05472216 -1.921076
alt 0.0000000 1.238390 0 4 0.05472216 -1.921076
$Cl3
Intern % glob % Intern freq Glob freq p.value v.test
secant 5.479452 1.857585 4 6 0.02772228 2.201194
neopre 2.739726 0.619195 2 2 0.05053555 1.955403
mantegós 2.739726 0.619195 2 2 0.05053555 1.955403
malaqualitat 2.739726 0.619195 2 2 0.05053555 1.955403
iode 2.739726 0.619195 2 2 0.05053555 1.955403
formatge 2.739726 0.619195 2 2 0.05053555 1.955403
cedre 2.739726 0.619195 2 2 0.05053555 1.955403
cartró 2.739726 0.619195 2 2 0.05053555 1.955403
astringent 5.479452 2.167183 4 7 0.05609819 1.910272
fusta 0.000000 2.786378 0 9 0.09643008 -1.662413
attr(,"class")
[1] "descfreq" "list " We only need to construct a vector containing the words that appear in one of the three groups.
Words characteristic of French judges in one of the three poles:
Words.Fig3.Fr <- unique(unlist(lapply(LC.Fig3.Fr$CharWord, rownames)))
Words.Fig3.Fr <- sort(Words.Fig3.Fr)
Words.Fig3.Fr
[1] "animal" "asséchant" "bois" "boisé" "charpenté" "dominé" "épice" "évent" "finale"
[10] "fraîcheur" "fruit" "fruité" "léger" "mûr" "neuf" "noir" "nonboisé" "rond"
[19] "souple" "tannique" "toasté" "très" "vanillé" Words characteristic of Catalan judges in one of the three poles:
Words.Fig3.Cat <- unique(unlist(lapply(LC.Fig3.Cat$CharWord, rownames)))
Words.Fig3.Cat <- sort(Words.Fig3.Cat)
Words.Fig3.Cat
[1] "alt" "astringent" "bota" "cafè" "cartró" "cedre" "claudolor"
[8] "confitura" "especiat" "formatge" "fruita" "fusta" "iode" "madur"
[15] "malaqualitat" "mantegós" "marcat" "neopre" "regalèssia" "secant" "sutja"
[22] "taní" "torrat"
The first 135 words are French:
Fr.Words.MFA <- rownames(res.mfact.23$quanti.var.sup$coord[1:135,])
Fr.Words.MFA
[1] "acétate" "acide" "acidité" "acidulé" "agréable" "agrume"
[7] "alcool" "amer" "amertume" "ample" "animal" "aspect"
[13] "asséchant" "astringence" "astringent" "attendre" "beau" "bois"
[19] "boisé" "bon" "bouche" "caramel" "carignan" "carton"
[25] "cassis" "cerise" "chair" "chaleureux" "charpenté" "complexe"
[31] "compoté" "concentration" "concentré" "confituré" "crayon" "cuit"
[37] "curieux" "cyste" "défaut" "degarde" "dominé" "doux"
[43] "dur" "emy" "encens" "épice" "épicé" "équilibre"
[49] "équilibré" "étable" "évent" "évolué" "évolution" "extrême"
[55] "farineux" "fin" "final" "finale" "floral" "fondu"
[61] "fort" "fraîche" "fraîcheur" "frais" "fruit" "fruité"
[67] "fumée" "garrigue" "généreux" "gouleyant" "gourmand" "gras"
[73] "grenache" "grillé" "humide" "jeune" "kirché" "lacté"
[79] "lactique" "léger" "limite" "long" "longueur" "manquedefondu"
[85] "matière" "minéral" "moins" "mou" "mûr" "mûre"
[91] "neuf" "neutre" "nez" "noir" "nonboisé" "olive"
[97] "onctueux" "particulier" "petitegarde" "peu" "peutannique" "plat"
[103] "plus" "prédominance" "présent" "prononcé" "pruneau" "puissance"
[109] "puissant" "rafle" "râpeux" "réduction" "réglisse" "résine"
[115] "rond" "rouge" "sec" "sécheresse" "souple" "structuré"
[121] "sucrée" "sucrosité" "taille" "tanin" "tannique" "toasté"
[127] "touche" "très" "trop" "type" "vanillé" "végétal"
[133] "velouté" "vif" "vin"
The words that occupy positions 136 to 230 are Catalan words:
Cat.Words.MFA <- rownames(res.mfact.23$freq.sup$coord[136:230,])
Cat.Words.MFA
[1] "acetatdetil" "acètic" "acidesabaix" "acidulat" "afrutat" "alcohòliques"
[7] "alt" "arbust" "aroma" "astringent" "balsàmic" "boca"
[13] "bota" "brisa" "cacau" "cafè" "caramel" "cartró"
[19] "cassis" "cedre" "cirera" "cítric" "claudolor" "complex"
[25] "compostos" "confitada" "confitura" "cos" "cosmitjabaix" "dens"
[31] "desequilibri" "dolç" "especiat" "espigola" "eucaliptus" "farigola"
[37] "floral" "florsseques" "formatge" "fruita" "fruitsec" "fum"
[43] "fumat" "fusta" "gerani" "glicerol" "greix" "herbesseques"
[49] "iode" "jove" "làctic" "liniment" "lleuger" "madur"
[55] "malaqualitat" "mantegós" "marcat" "mel" "melós" "mentolat"
[61] "mora" "nas" "neopre" "oxidat" "pebrotverd" "pegadolça"
[67] "picant" "picat" "pinassa" "pla" "pocaaroma" "pocestructurat"
[73] "pocpersistent" "pollen" "potència" "prunasec" "químic" "regalèssia"
[79] "rodó" "roure" "secant" "sensacions" "sofre" "sotabosc"
[85] "sucrositat" "sutja" "taní" "tànnic" "toffe" "torrat"
[91] "torrefacte" "vainilla" "vegetal" "vi" "xocolata"
To build a vector with the selected characteristic words and plot them:
pos.words.Fig3.Fr <- which(Fr.Words.MFA %in% Words.Fig3.Fr)
pos.words.Fig3.Cat <- which(Cat.Words.MFA %in% Words.Fig3.Cat)
pos.words.Fig3 <- c(pos.words.Fig3.Fr, 135+pos.words.Fig3.Cat)
plot.MFA(res.mfact.23,choix=c("freq"),invisible=c("row","col"),axes=c(1,2), select= pos.words.Fig3,unselect=1,
legend=list(plot=FALSE),habillage="none",autoLab = c("yes"),cex=0.8, title="",graph.type="classic")
Characteristic words in English on the MFACT plane using different colors for each judge group
Separate translations are provided for French and Catalan words.
Characteristic French words
Coord.Fr <- data.frame(res.mfact.23$freq.sup$coord[pos.words.Fig3.Fr,], lang="Fr")
Coord.Fr
Dim.1 Dim.2 Dim.3 Dim.4 Dim.5 Dim.6 Dim.7 lang
animal -0.59343967 -0.511722034 -0.290823763 -0.14090781 -0.14469656 -5.900284e-01 -0.36933291 Fr
asséchant -0.67338104 -0.803561875 -0.101463172 0.27194795 0.26204007 -9.519306e-02 0.06850627 Fr
bois 1.30114701 0.051069203 -0.483807494 0.08205348 -0.22732622 4.965811e-01 -0.19931170 Fr
boisé 0.66641435 -0.480026128 -0.044177786 0.10843598 -0.03863853 1.937070e-01 0.13171020 Fr
charpenté 1.24802969 -0.336910906 -0.255218575 0.22412317 -0.01192454 2.704951e-01 0.26337581 Fr
dominé 1.48972280 -0.132100595 -0.425383218 0.44185159 -0.56820124 6.596303e-01 -0.27950682 Fr
épice -0.01134125 0.253361545 -0.831494258 0.18607987 -0.17130206 -2.950341e-01 0.16920614 Fr
évent -0.92317035 0.072908887 -0.754317378 0.28753767 0.01821749 -3.888060e-02 -0.09357353 Fr
finale 1.23519276 -0.323182362 -0.272182884 0.14037764 0.37991193 -5.370408e-03 0.47145789 Fr
fraîcheur -0.60089901 0.385991015 0.172514233 0.19735824 0.05102441 8.928621e-02 0.34090288 Fr
fruit -0.47749854 0.355953911 -0.027557246 0.06103203 -0.14054131 3.862215e-02 -0.06106851 Fr
fruité -0.39982720 0.329897705 0.288711006 -0.08539560 0.19727933 -4.020929e-05 -0.09620369 Fr
léger -0.84266652 0.315674093 -0.285823379 0.17022403 0.44263986 7.037213e-01 0.32350305 Fr
mûr -0.32364408 0.302200189 -0.019685111 -0.63726117 -0.17254803 1.266075e-01 -0.08732380 Fr
neuf 1.58489551 -0.002238351 -0.544394158 0.38322473 -0.06266666 3.024667e-01 -0.13478398 Fr
noir -0.51200080 0.979784095 0.588623369 0.08993399 0.22809222 3.054174e-01 -0.03684067 Fr
nonboisé -0.68346651 0.164631867 0.048985206 0.30313197 0.20450185 1.163886e-01 0.06586744 Fr
rond -0.51200080 0.979784095 0.588623369 0.08993399 0.22809222 3.054174e-01 -0.03684067 Fr
souple -0.03439260 0.902010962 0.328552340 -0.57750861 0.04507460 -1.428955e-01 -0.04713331 Fr
tannique 0.30973816 -0.467442131 -0.005427685 0.10256343 -0.31188564 -2.826680e-01 0.08310495 Fr
toasté 0.29012704 -1.118934073 0.300860299 -0.16743876 0.63777844 -2.796487e-01 0.47369343 Fr
très 0.33371388 0.531209809 -0.227265792 -0.34394285 0.07777862 1.705868e-01 -0.05254504 Fr
vanillé 0.70680948 -0.354176022 -0.232543146 -0.06772137 -0.05969652 4.114211e-02 0.08403530 FrCharacteristic Catalan words
Coord.Cat <- data.frame(res.mfact.23$freq.sup$coord[135+pos.words.Fig3.Cat,], lang="Cat")
Coord.Cat
Dim.1 Dim.2 Dim.3 Dim.4 Dim.5 Dim.6 Dim.7 lang
alt 1.5947624 0.036211572 -0.4229721 0.35593743 -0.07250461 0.31522990 -0.101814243 Cat
astringent 0.2729478 -0.671920946 0.5839807 0.03244007 0.23168459 -0.08569083 -0.098122551 Cat
bota 1.5947624 0.036211572 -0.4229721 0.35593743 -0.07250461 0.31522990 -0.101814243 Cat
cafè 1.4995897 -0.093650671 -0.3039612 0.41456430 -0.57803919 0.67239344 -0.246537076 Cat
cartró -0.2953690 -1.555291851 0.7231142 -0.25969536 0.44322841 -0.23332663 -0.097343151 Cat
cedre -0.2953690 -1.555291851 0.7231142 -0.25969536 0.44322841 -0.23332663 -0.097343151 Cat
claudolor 1.1700245 0.337620046 -0.4803152 -0.33434536 0.35647814 0.16771341 -0.013785433 Cat
confitura -0.4074681 0.108617159 0.4205172 -0.35982503 -0.01193924 -0.14860409 -0.007032928 Cat
especiat 1.1978684 0.199421005 -0.3974400 -0.16111268 -0.03263916 0.41151483 -0.118224895 Cat
formatge -0.2953690 -1.555291851 0.7231142 -0.25969536 0.44322841 -0.23332663 -0.097343151 Cat
fruita -0.4190979 0.443405443 0.4582364 -0.11395668 -0.22491157 -0.36188649 0.048398249 Cat
fusta 1.2179704 0.326736013 -0.1746441 0.07115059 0.12553131 0.18164269 -0.061806163 Cat
iode -0.2953690 -1.555291851 0.7231142 -0.25969536 0.44322841 -0.23332663 -0.097343151 Cat
madur -0.2837087 0.006574687 0.6378617 -0.53775737 -0.02690668 -0.13696810 0.190794988 Cat
malaqualitat -0.2953690 -1.555291851 0.7231142 -0.25969536 0.44322841 -0.23332663 -0.097343151 Cat
mantegós -0.2953690 -1.555291851 0.7231142 -0.25969536 0.44322841 -0.23332663 -0.097343151 Cat
marcat 1.5947624 0.036211572 -0.4229721 0.35593743 -0.07250461 0.31522990 -0.101814243 Cat
neopre -0.2953690 -1.555291851 0.7231142 -0.25969536 0.44322841 -0.23332663 -0.097343151 Cat
regalèssia 0.6234730 -0.376522195 -0.1704102 0.32518356 0.19355106 0.26633032 -0.035120850 Cat
secant -0.5367664 -0.947644286 0.1811539 0.05678312 0.16783263 -0.27450917 0.059039122 Cat
sutja 1.5947624 0.036211572 -0.4229721 0.35593743 -0.07250461 0.31522990 -0.101814243 Cat
taní 0.6496967 -0.759540140 0.1500710 0.04812104 0.18536190 0.04095164 -0.099578697 Cat
torrat 1.1421585 -0.243802765 -0.3006566 0.20302783 0.50108518 -0.18695654 -0.488736316 CatFrench words:
Words.Fig3.Fr
[1] "animal" "asséchant" "bois" "boisé" "charpenté" "dominé"
[7] "épice" "évent" "finale" "fraîcheur" "fruit" "fruité"
[13] "léger" "mûr" "neuf" "noir" "nonboisé" "rond"
[19] "souple" "tannique" "toasté" "très" "vanillé" df.Char.Fr <- data.frame(orig= "animal", transl="animal")
df.Char.Fr <- rbind(df.Char.Fr, data.frame(orig= "asséchant", transl="drying") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "bois", transl="wood") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "boisé", transl="woody") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "charpenté", transl="structured") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "dominé", transl="surpass") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "épice", transl="spicy") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "évent", transl="staleness") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "finale", transl="final") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "fraîcheur", transl="freshness") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "fruit", transl="fruit") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "fruité", transl="fruity") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "léger", transl="light") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "mûr", transl="ripe") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "neuf", transl="new") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "noir", transl="black") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "nonboisé", transl="unwooded") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "rond", transl="round") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "souple", transl="supple") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "tannique", transl="tannic") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "toasté", transl="toasted") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "très", transl="very") )
df.Char.Fr<- rbind(df.Char.Fr, data.frame(orig= "vanillé", transl="vanillin") )
df.Char.Fr
orig transl
1 animal animal
2 asséchant drying
3 bois wood
4 boisé woody
5 charpenté structured
6 dominé surpass
7 épice spicy
8 évent staleness
9 finale final
10 fraîcheur freshness
11 fruit fruit
12 fruité fruity
13 léger light
14 mûr ripe
15 neuf new
16 noir black
17 nonboisé unwooded
18 rond round
19 souple supple
20 tannique tannic
21 toasté toasted
22 très very
23 vanillé vanillin
To join French coordinates and their translation:
Coord.Fr.Fig3 <- merge(Coord.Fr, df.Char.Fr, by.x=0, by.y="orig")
Coord.Fr.Fig3[,c(1:5,9,10)]
Row.names Dim.1 Dim.2 Dim.3 Dim.4 lang transl
1 animal -0.59343967 -0.511722034 -0.290823763 -0.14090781 Fr animal
2 asséchant -0.67338104 -0.803561875 -0.101463172 0.27194795 Fr drying
3 bois 1.30114701 0.051069203 -0.483807494 0.08205348 Fr wood
4 boisé 0.66641435 -0.480026128 -0.044177786 0.10843598 Fr woody
5 charpenté 1.24802969 -0.336910906 -0.255218575 0.22412317 Fr structured
6 dominé 1.48972280 -0.132100595 -0.425383218 0.44185159 Fr surpass
7 épice -0.01134125 0.253361545 -0.831494258 0.18607987 Fr spicy
8 évent -0.92317035 0.072908887 -0.754317378 0.28753767 Fr staleness
9 finale 1.23519276 -0.323182362 -0.272182884 0.14037764 Fr final
10 fraîcheur -0.60089901 0.385991015 0.172514233 0.19735824 Fr freshness
11 fruit -0.47749854 0.355953911 -0.027557246 0.06103203 Fr fruit
12 fruité -0.39982720 0.329897705 0.288711006 -0.08539560 Fr fruity
13 léger -0.84266652 0.315674093 -0.285823379 0.17022403 Fr light
14 mûr -0.32364408 0.302200189 -0.019685111 -0.63726117 Fr ripe
15 neuf 1.58489551 -0.002238351 -0.544394158 0.38322473 Fr new
16 noir -0.51200080 0.979784095 0.588623369 0.08993399 Fr black
17 nonboisé -0.68346651 0.164631867 0.048985206 0.30313197 Fr unwooded
18 rond -0.51200080 0.979784095 0.588623369 0.08993399 Fr round
19 souple -0.03439260 0.902010962 0.328552340 -0.57750861 Fr supple
20 tannique 0.30973816 -0.467442131 -0.005427685 0.10256343 Fr tannic
21 toasté 0.29012704 -1.118934073 0.300860299 -0.16743876 Fr toasted
22 très 0.33371388 0.531209809 -0.227265792 -0.34394285 Fr very
23 vanillé 0.70680948 -0.354176022 -0.232543146 -0.06772137 Fr vanillinCatalan words:
Words.Fig3.Cat
[1] "alt" "astringent" "bota" "cafè" "cartró"
[6] "cedre" "claudolor" "confitura" "especiat" "formatge"
[11] "fruita" "fusta" "iode" "madur" "malaqualitat"
[16] "mantegós" "marcat" "neopre" "regalèssia" "secant"
[21] "sutja" "taní" "torrat" df.Char.Cat <- data.frame(orig= "alt", transl="high")
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "astringent", transl="astringent") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "bota", transl="barrel") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "cafè", transl="coffee") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "cartró", transl="cardboard") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "cedre", transl="cedar") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "claudolor", transl="clove") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "confitura", transl="jelly") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "especiat", transl="spicy") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "formatge", transl="cheese") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "fruita", transl="fruit") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "fusta", transl="wood") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "iode", transl="iodine") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "madur", transl="mature") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "malaqualitat", transl="poorquality") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "mantegós", transl="buttered") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "marcat", transl="marked") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "neopre", transl="neoprene") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "regalèssia", transl="licorice") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "secant", transl="drying") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "sutja", transl="soot") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "taní", transl="tannin") )
df.Char.Cat <- rbind(df.Char.Cat , data.frame(orig= "torrat", transl="toasted") )
df.Char.Cat
orig transl
1 alt high
2 astringent astringent
3 bota barrel
4 cafè coffee
5 cartró cardboard
6 cedre cedar
7 claudolor clove
8 confitura jelly
9 especiat spicy
10 formatge cheese
11 fruita fruit
12 fusta wood
13 iode iodine
14 madur mature
15 malaqualitat poorquality
16 mantegós buttered
17 marcat marked
18 neopre neoprene
19 regalèssia licorice
20 secant drying
21 sutja soot
22 taní tannin
23 torrat toasted
To join Catalan coordinates and their translation:
Coord.Cat.Fig3 <- merge(Coord.Cat, df.Char.Cat, by.x=0, by.y="orig")
Coord.Cat.Fig3[,c(1:5,9,10)]
Row.names Dim.1 Dim.2 Dim.3 Dim.4 lang transl
1 alt 1.5947624 0.036211572 -0.4229721 0.35593743 Cat high
2 astringent 0.2729478 -0.671920946 0.5839807 0.03244007 Cat astringent
3 bota 1.5947624 0.036211572 -0.4229721 0.35593743 Cat barrel
4 cafè 1.4995897 -0.093650671 -0.3039612 0.41456430 Cat coffee
5 cartró -0.2953690 -1.555291851 0.7231142 -0.25969536 Cat cardboard
6 cedre -0.2953690 -1.555291851 0.7231142 -0.25969536 Cat cedar
7 claudolor 1.1700245 0.337620046 -0.4803152 -0.33434536 Cat clove
8 confitura -0.4074681 0.108617159 0.4205172 -0.35982503 Cat jelly
9 especiat 1.1978684 0.199421005 -0.3974400 -0.16111268 Cat spicy
10 formatge -0.2953690 -1.555291851 0.7231142 -0.25969536 Cat cheese
11 fruita -0.4190979 0.443405443 0.4582364 -0.11395668 Cat fruit
12 fusta 1.2179704 0.326736013 -0.1746441 0.07115059 Cat wood
13 iode -0.2953690 -1.555291851 0.7231142 -0.25969536 Cat iodine
14 madur -0.2837087 0.006574687 0.6378617 -0.53775737 Cat mature
15 malaqualitat -0.2953690 -1.555291851 0.7231142 -0.25969536 Cat poorquality
16 mantegós -0.2953690 -1.555291851 0.7231142 -0.25969536 Cat buttered
17 marcat 1.5947624 0.036211572 -0.4229721 0.35593743 Cat marked
18 neopre -0.2953690 -1.555291851 0.7231142 -0.25969536 Cat neoprene
19 regalèssia 0.6234730 -0.376522195 -0.1704102 0.32518356 Cat licorice
20 secant -0.5367664 -0.947644286 0.1811539 0.05678312 Cat drying
21 sutja 1.5947624 0.036211572 -0.4229721 0.35593743 Cat soot
22 taní 0.6496967 -0.759540140 0.1500710 0.04812104 Cat tannin
23 torrat 1.1421585 -0.243802765 -0.3006566 0.20302783 Cat toasted
To build Figure 3:
ax1 <-1 ; ax2 <-2
labx <- paste0("Dim 1 (", round(res.mfact.23$eig[ax1,2],1),"%)" )
laby <- paste0("Dim 2 (", round(res.mfact.23$eig[ax2,2],1),"%)" )
font.type.FRCat <- c("bold", "bold.italic")
color.type.FRCat <- c("Green", "Blue")col.margin = c("black", "red")
Coord.Fig3$lang <- as.factor(Coord.Fig3$lang)
set.seed(123)
library(ggtext)
Figure3 <- ggplot(Coord.Fig3)+
theme_light() + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())+
xlab(labx)+ ylab(laby) + coord_fixed()+
geom_hline(yintercept=0, linetype="dashed", color = "grey")+
geom_vline(xintercept=0, linetype="dashed", color = "grey")+
geom_text_repel(size=5, fontface = font.type.FRCat[Coord.Fig3$lang], max.overlaps=100,
box.padding = unit(0.35, "lines"),
aes(x=Dim.1, y=Dim.2, label = transl, color=lang))+
theme( axis.text.x = element_text(size=rel(1.6)))+
theme( axis.text.y = element_text(size=rel(1.6)))+
labs(x=labx)+labs(y=laby)+
theme(axis.title.x= element_text(size=17, face="bold"))+
theme(axis.title.y= element_text(size=17, face="bold"))+
theme(plot.margin = grid::unit(c(t=5,r= 2,b=5, l=2), "mm"))+
scale_color_manual(name="Language",
labels=c("Catalan","French"),
values = setNames(col.margin, levels(Coord.Fig3$lang))) +
theme(axis.title.x = element_text(margin=margin(t=10))) +
theme(panel.border = element_rect(colour = "black", fill=NA, linewidth=1)) +
theme(legend.position = "none")+
labs(title = "Words originally in <b style='color:#FF0000'>**_French_**</b> and **Catalan**")+
theme(plot.title = ggtext::element_markdown(lineheight = 1.1, hjust=1, size=20))+
ylim(-1.9, 1.1)+
ggtitle("Figure 3. Characteristic words of the three poles")
To plot Figure 3:
set.seed(123)
Figure3
