Pregunta 1

¿Cómo han cambiado los tópicos de interés de las áreas de conocimiento durante cierto período de tiempo?

#Librerías
library(quanteda)
library(tidyr)
library(dplyr)
library(tidyverse)
library(quanteda.textstats)
library(quanteda.textplots)
library(pluralize)
datos_uv = read.csv("C:/Users/ureus/Downloads/ICB_TM - BD UV.csv", sep=",")
tesauro = read.csv("C:/Users/ureus/Downloads/IEEE_tesauro.csv", sep=";")

#Limpieza y edición del dataset
datos_uv$Index.keywords = tolower(datos_uv$Index.keywords)
tesauro$Area2 = str_trim(tesauro$Area2)
tesauro$Area2 = gsub(" ", "_", tesauro$Area2)
tesauro$Area2 = tolower(tesauro$Area2)
#Arreglo del dataset
datos_uv <- datos_uv%>%
  mutate(Index.keywords = strsplit(as.character(Index.keywords),";"))%>%
  unnest(Index.keywords)
#Eliminar espacios en blanco y juntar palabras separados por ellos.
datos_uv$Index.keywords = str_trim(datos_uv$Index.keywords)
datos_uv$Index.keywords = gsub(" ", "_", datos_uv$Index.keywords)
datos_uv$Index.keywords = singularize(datos_uv$Index.keywords)

#Eliminar palabras innecesarias
palabras_innecesarias<-c(tm::stopwords(kind="en")," human"," female"," male"," humans"," adolescents",
                         " review"," prospective study"," article"," human"," young adult",
                         " controlled study", " priority_journal", "young_adult", " human",
                         "female","male","humans","adolescents", "review","prospective study",
                         "article","human","young adult", "adult", " adult",
                         "controlled study", "priority journal", "young adult", "human", 
                         "middle aged", " middle aged", " aged", "aged", "Aged", "adolescent", " adolescent",
                         "clinical article", " clinical article", "Older_adults",
                         "Non-intrusive", "child", "chilean", "chile","clinical_article",
                         "controlled_study","human_experiment","normal_human",
                         "rat","nonhuman","animal","wistar_rat","animals","infant",
                         "priority_journal","middle_age", "middle_aged", "animal_experiment")

datos_uv <- datos_uv %>%
  filter(!Index.keywords %in% palabras_innecesarias) %>%
  filter(SI =="SI" )
#Creación del corpus
corpus_datos_uv = corpus(datos_uv, text_field = "Index.keywords")

#Tokenización 
toks_datos_uv = tokens(corpus_datos_uv,remove_punct = T,remove_symbols = T,
                       remove_numbers = T, remove_url = T,
                    remove_separators = T)
#Convertir a minúsculas
toks_datos_uv=tokens_tolower(toks_datos_uv)

uv_datos <- datos_uv %>% 
  inner_join(tesauro, by=c("Index.keywords" = "Area2"))

#Transformar en corpus
corpus_uv = corpus(uv_datos, text_field="Index.keywords")
corpus_uv = tokens(corpus_uv)

dfmat_datos_uv = dfm(corpus_uv)
#Graficar
ano = "2021"
acumulado = F

if (acumulado==F){
  tstat_key = textstat_keyness(dfmat_datos_uv,target = dfmat_datos_uv$Año == ano)} else
  {tstat_key = textstat_keyness(dfmat_datos_uv,target = dfmat_datos_uv$Año <= ano)}

grafico_area=textplot_keyness(tstat_key,labelsize = 4,n=10,margin = 0.6, 
                                   color = c("mediumaquamarine", "gray"))
plot(grafico_area)