Comparación Guion Aladdin 1992-2019

Introducción

La actividad corresponde en comparar dos guiones de película, uno es de Aladdin de 1992 y el otro es de Aladdin de 2019. El primero es una versión animada, por lo cual la tarea es analizar que tanto se ha modernizado el nuevo guión y si sigue siendo fiel a su antecesor.

library(tm)

## Loading required package: NLP

library(pdftools)

## Using poppler version 23.04.0

library(SnowballC)
library(NLP)
library(wordcloud)

## Loading required package: RColorBrewer

library(RColorBrewer)
library(syuzhet)

A. Depuración, Limpieza y estructura de Texto

• Buscar
• Limpiar
• Estructurar

Aladdin 1992 (.txt)

Sys.setenv(LANG = "en")

#Let's start by importing the text to analyze
tex <- readLines(("Aladdin1992.txt"), warn = FALSE)
text <- iconv(tex, "WINDOWS-1252", "UTF-8")

#We need to convert the text to a corpus

docs <- Corpus(VectorSource(text))
#inspect(docs)

#General text cleaning

# Convert the text to lower case
docs <- tm_map(docs, content_transformer(tolower))

## Warning in tm_map.SimpleCorpus(docs, content_transformer(tolower)):
## transformation drops documents

# Remove numbers
docs <- tm_map(docs, removeNumbers)

## Warning in tm_map.SimpleCorpus(docs, removeNumbers): transformation drops
## documents

# Remove english common stopwords
docs <- tm_map(docs, removeWords, stopwords("english"))

## Warning in tm_map.SimpleCorpus(docs, removeWords, stopwords("english")):
## transformation drops documents

# Remove your own stop word
# specify your stopwords as a character vector
docs <- tm_map(docs, removeWords, c("blabla1", "blabla2")) 

## Warning in tm_map.SimpleCorpus(docs, removeWords, c("blabla1", "blabla2")):
## transformation drops documents

# Remove punctuations
docs <- tm_map(docs, removePunctuation)

## Warning in tm_map.SimpleCorpus(docs, removePunctuation): transformation drops
## documents

# Eliminate extra white spaces
docs <- tm_map(docs, stripWhitespace) 

## Warning in tm_map.SimpleCorpus(docs, stripWhitespace): transformation drops
## documents

#Term-document matrix. Document matrix is a table containing the frequency of the words.

dtm <- TermDocumentMatrix(docs)
m <- as.matrix(dtm)
v <- sort(rowSums(m),decreasing=TRUE)
d <- data.frame(word = names(v),freq=v)
head(d, 100)

##              word freq
## aladdin   aladdin  382
## jafar       jafar  200
## jasmine   jasmine  194
## abu           abu  168
## genie       genie  166
## sultan     sultan  111
## iago         iago   93
## carpet     carpet   89
## back         back   72
## prince     prince   63
## looks       looks   60
## lamp         lamp   57
## like         like   52
## one           one   48
## princess princess   46
## head         head   44
## just         just   44
## see           see   43
## turns       turns   42
## pulls       pulls   41
## ali           ali   40
## look         look   38
## now           now   38
## get           get   35
## got           got   35
## sees         sees   35
## right       right   32
## will         will   32
## man           man   32
## away         away   32
## can           can   31
## never       never   30
## wish         wish   30
## cave         cave   29
## guards     guards   29
## make         make   29
## cut           cut   28
## guard       guard   28
## grabs       grabs   28
## well         well   28
## boy           boy   27
## rajah       rajah   27
## hand         hand   26
## around     around   26
## come         come   25
## good         good   25
## comes       comes   25
## way           way   25
## yes           yes   24
## begins     begins   24
## aladdins aladdins   24
## palace     palace   24
## old           old   24
## hey           hey   23
## little     little   22
## two           two   22
## new           new   22
## top           top   21
## want         want   21
## face         face   20
## know         know   20
## hands       hands   20
## street     street   20
## time         time   20
## gotta       gotta   20
## think       think   19
## flies       flies   19
## goes         goes   19
## genies     genies   19
## world       world   19
## falls       falls   18
## normal     normal   18
## jumps       jumps   18
## free         free   18
## mouth       mouth   17
## gonna       gonna   17
## take         take   17
## takes       takes   17
## walks       walks   17
## staff       staff   17
## say           say   17
## whole       whole   17
## finally   finally   16
## jafars     jafars   16
## three       three   16
## tell         tell   16
## door         door   16
## appears   appears   16
## pile         pile   15
## turban     turban   15
## stop         stop   14
## another   another   14
## friend     friend   14
## find         find   14
## eyes         eyes   14
## rat           rat   14
## runs         runs   14
## love         love   14
## sorry       sorry   14
## floor       floor   14

findFreqTerms(dtm, lowfreq = 20)

##  [1] "aladdin"  "come"     "cut"      "face"     "like"     "hey"     
##  [7] "back"     "right"    "carpet"   "good"     "guard"    "little"  
## [13] "look"     "yes"      "will"     "pulls"    "never"    "one"     
## [19] "begins"   "can"      "see"      "lamp"     "man"      "hand"    
## [25] "iago"     "jafar"    "comes"    "get"      "grabs"    "two"     
## [31] "head"     "cave"     "now"      "away"     "know"     "turns"   
## [37] "just"     "got"      "hands"    "street"   "looks"    "around"  
## [43] "top"      "way"      "aladdins" "abu"      "time"     "guards"  
## [49] "gotta"    "make"     "wish"     "sees"     "prince"   "palace"  
## [55] "princess" "boy"      "sultan"   "jasmine"  "rajah"    "want"    
## [61] "well"     "old"      "genie"    "ali"      "new"

Sys.setenv(LANG = "en")

Aladdin 2019 (.txt)

#Let's start by importing the text to analyze
tex2 <- readLines(("Aladdin_2019.txt"), warn=FALSE)
text2 <- iconv(tex2,"WINDOWS-1252","UTF-8")

#We need to convert the text to a corpus

docs2 <- Corpus(VectorSource(text))
#inspect(docs)

#General text cleaning

# Convert the text to lower case
docs2 <- tm_map(docs, content_transformer(tolower))

## Warning in tm_map.SimpleCorpus(docs, content_transformer(tolower)):
## transformation drops documents

# Remove numbers
docs2 <- tm_map(docs, removeNumbers)

## Warning in tm_map.SimpleCorpus(docs, removeNumbers): transformation drops
## documents

# Remove english common stopwords
docs2 <- tm_map(docs, removeWords, stopwords("english"))

## Warning in tm_map.SimpleCorpus(docs, removeWords, stopwords("english")):
## transformation drops documents

# Remove your own stop word
# specify your stopwords as a character vector
docs2 <- tm_map(docs, removeWords, c("blabla1", "blabla2")) 

## Warning in tm_map.SimpleCorpus(docs, removeWords, c("blabla1", "blabla2")):
## transformation drops documents

# Remove punctuations
docs2 <- tm_map(docs, removePunctuation)

## Warning in tm_map.SimpleCorpus(docs, removePunctuation): transformation drops
## documents

# Eliminate extra white spaces
docs2 <- tm_map(docs, stripWhitespace) 

## Warning in tm_map.SimpleCorpus(docs, stripWhitespace): transformation drops
## documents

#Term-document matrix. Document matrix is a table containing the frequency of the words.

dtm2 <- TermDocumentMatrix(docs2)
m2 <- as.matrix(dtm2)
v2 <- sort(rowSums(m2),decreasing=TRUE)
d2 <- data.frame(word = names(v2),freq=v2)
head(d2, 100)

##              word freq
## aladdin   aladdin  382
## jafar       jafar  200
## jasmine   jasmine  194
## abu           abu  168
## genie       genie  166
## sultan     sultan  111
## iago         iago   93
## carpet     carpet   89
## back         back   72
## prince     prince   63
## looks       looks   60
## lamp         lamp   57
## like         like   52
## one           one   48
## princess princess   46
## head         head   44
## just         just   44
## see           see   43
## turns       turns   42
## pulls       pulls   41
## ali           ali   40
## look         look   38
## now           now   38
## get           get   35
## got           got   35
## sees         sees   35
## right       right   32
## will         will   32
## man           man   32
## away         away   32
## can           can   31
## never       never   30
## wish         wish   30
## cave         cave   29
## guards     guards   29
## make         make   29
## cut           cut   28
## guard       guard   28
## grabs       grabs   28
## well         well   28
## boy           boy   27
## rajah       rajah   27
## hand         hand   26
## around     around   26
## come         come   25
## good         good   25
## comes       comes   25
## way           way   25
## yes           yes   24
## begins     begins   24
## aladdins aladdins   24
## palace     palace   24
## old           old   24
## hey           hey   23
## little     little   22
## two           two   22
## new           new   22
## top           top   21
## want         want   21
## face         face   20
## know         know   20
## hands       hands   20
## street     street   20
## time         time   20
## gotta       gotta   20
## think       think   19
## flies       flies   19
## goes         goes   19
## genies     genies   19
## world       world   19
## falls       falls   18
## normal     normal   18
## jumps       jumps   18
## free         free   18
## mouth       mouth   17
## gonna       gonna   17
## take         take   17
## takes       takes   17
## walks       walks   17
## staff       staff   17
## say           say   17
## whole       whole   17
## finally   finally   16
## jafars     jafars   16
## three       three   16
## tell         tell   16
## door         door   16
## appears   appears   16
## pile         pile   15
## turban     turban   15
## stop         stop   14
## another   another   14
## friend     friend   14
## find         find   14
## eyes         eyes   14
## rat           rat   14
## runs         runs   14
## love         love   14
## sorry       sorry   14
## floor       floor   14

findFreqTerms(dtm, lowfreq = 20)

##  [1] "aladdin"  "come"     "cut"      "face"     "like"     "hey"     
##  [7] "back"     "right"    "carpet"   "good"     "guard"    "little"  
## [13] "look"     "yes"      "will"     "pulls"    "never"    "one"     
## [19] "begins"   "can"      "see"      "lamp"     "man"      "hand"    
## [25] "iago"     "jafar"    "comes"    "get"      "grabs"    "two"     
## [31] "head"     "cave"     "now"      "away"     "know"     "turns"   
## [37] "just"     "got"      "hands"    "street"   "looks"    "around"  
## [43] "top"      "way"      "aladdins" "abu"      "time"     "guards"  
## [49] "gotta"    "make"     "wish"     "sees"     "prince"   "palace"  
## [55] "princess" "boy"      "sultan"   "jasmine"  "rajah"    "want"    
## [61] "well"     "old"      "genie"    "ali"      "new"

B. Análisis del Texto

• Frecuencias
• Comparación
• Asociaciones

Gráficas de barra

barplot(d[1:10,]$freq, las = 2, names.arg = d[1:10,]$word,
        col ="orange", main ="Palabras más frecuentes - Aladdin 1992",
        ylab = "Frecunecia de palabras")

head(d)

##            word freq
## aladdin aladdin  382
## jafar     jafar  200
## jasmine jasmine  194
## abu         abu  168
## genie     genie  166
## sultan   sultan  111

barplot(d2[1:10,]$freq, las = 2, names.arg = d2[1:10,]$word,
        col ="brown", main ="Palabras más frecuentes - Aladdin 2019",
        ylab = "Frecunecia de palabras")

head(d2)

##            word freq
## aladdin aladdin  382
## jafar     jafar  200
## jasmine jasmine  194
## abu         abu  168
## genie     genie  166
## sultan   sultan  111

Wordcloud

Aladdin 1992

set.seed(1234)
wordcloud(words = d$word, freq = d$freq, min.freq = 15,
          max.words=Inf, random.order=T, rot.per=0.5, 
          colors=brewer.pal(8, "Dark2"))

Aladdin 2019

wordcloud(words = d2$word, freq = d2$freq, min.freq = 15,
          max.words=Inf, random.order=T, rot.per=0.5, 
          colors=brewer.pal(8, "Dark2"))

Asociaciones

Aladdin 1992

findAssocs(dtm, terms = "aladdins", corlimit = 0.2)

## $aladdins
## shoulder  stopped    trips    bulge     name    tools   debris  diploma 
##     0.23     0.20     0.20     0.20     0.20     0.20     0.20     0.20 
##    slaps   arabic     list    hopin   widely 
##     0.20     0.20     0.20     0.20     0.20

Aladdin 2019

findAssocs(dtm2, terms = "aladdins", corlimit = 0.2)

## $aladdins
## shoulder  stopped    trips    bulge     name    tools   debris  diploma 
##     0.23     0.20     0.20     0.20     0.20     0.20     0.20     0.20 
##    slaps   arabic     list    hopin   widely 
##     0.20     0.20     0.20     0.20     0.20

C. Preguntas

1. Además de limpiar contesta a esta pregunta ¿Se tomará y analizará Todo lo que aparece en la página? ¿Que si y que no? ¿Por qué?

No se toma en cuenta todo el texto del archivo, aquelas secciones descriptivas, instrucciones de escenas o especificaciones técnicas quedan fuera del análisis.

2. Hacer una comparación de frecuencia y un análisis de asociasión muy básico.

El análisis de asociación para “aladdins” en ambos guiones de “Aladdin” muestra una notable consistencia, con términos como “shoulder”, “stopped”, y “arabic” asociados estrechamente. Esta similitud puede sugiere que ciertos aspectos del personaje y elementos culturales permanecen constantes entre las versiones de 1992 y 2019.

D. Funcion Nueva - Análisis de Sentimiento

Busca alguna función asociada al análisis de texto que no hayamos visto en clase, úsala y escribe lo que lograste (aquí hay muchas funciones posibles, con que sea algo no visto en clase está bien)

aladdin_1992 <- readLines("Aladdin1992.txt", warn = FALSE)
aladdin_1992 <- iconv(aladdin_1992, "WINDOWS-1252", "UTF-8")
aladdin_2019 <- readLines("Aladdin_2019.txt", warn = FALSE)
aladdin_2019 <- iconv(aladdin_2019, "WINDOWS-1252", "UTF-8")

# Create corpora
corpus_1992 <- Corpus(VectorSource(aladdin_1992))
corpus_2019 <- Corpus(VectorSource(aladdin_2019))

# Análisis de sentimientos para Aladdin 1992
sentiments_1992 <- get_sentiment(as.character(corpus_1992), method = "syuzhet")

# Análisis de sentimientos para Aladdin 2019
sentiments_2019 <- get_sentiment(as.character(corpus_2019), method = "syuzhet")

# Calculamos la media de sentimientos para obtener una visión general del tono de cada guion de Aladdin
mean_sentiment_1992 <- mean(sentiments_1992)
mean_sentiment_2019 <- mean(sentiments_2019)

# Mostramos los resultados
cat("Media de sentimientos para Aladdin 1992:", mean_sentiment_1992, "\n")

## Media de sentimientos para Aladdin 1992: -8.983333

cat("Media de sentimientos para Aladdin 2019:", mean_sentiment_2019)

## Media de sentimientos para Aladdin 2019: 12.31667

Resultados

• Tono Más Positivo en 2019: La media positiva del guion de 2019 sugiere un enfoque más optimista en comparación con la versión de 1992, posiblemente destinado a hacer la historia más inspiradora, esperanzadora o alegre.

• Adaptación a la Audiencia Actual: La variación en el tono refleja las preferencias cambiantes de la audiencia, que busca historias que transmitan mensajes positivos de esperanza y resiliencia.

• Impacto de la Cultura Popular y Sensibilidades Modernas: La versión de 2019 puede estar influenciada por las sensibilidades actuales hacia la inclusión y el optimismo frente a la adversidad.

La diferencia entre los dos guiones representan sin duda alguna una evoluci´øn en la forma de contar historias, se puede intuir que el optimismo en el cine está en crecimiento en historias clásicas, cambiando su narrativa, personajes, tonos, etc.