R/Syn&Ant.R

In Fidelalencar/LCtestcontentR: Functionalities and objects for content production to the website.

##################################
### (1) criando função que executa as sugestões de sinonimos


# example <- "https://www.youtube.com/watch?v=4GsjEPRMzdw" # easy
# example <- "https://www.youtube.com/watch?v=LsFdROZ2OdA" # difficult
# Legendas <- youtubecaption::get_caption(url = example,
#                                         language = "en",  # ATENCAO PARA A ESCOLHA DA LEGENDA
#                                         savexl = FALSE, openxl = FALSE, path = getwd())
#
# Legendas_selec <- Legendas$text[20:140]
# LegendasU <- paste(Legendas_selec, collapse= " ")
# LegendasU
#
# LL <- our_tokenizer(LegendasU)
# LL
#
# testdata_Level <- LL
#
# # setwd("C:/Users/Paren/Dropbox/Learning_Community/LCtestcontentR/data")
# # load(file="testdata_Level.RData")
# # load(file="Syn.1.0.Rdata")
# # testdata_Level <- testdata_Level[["easy"]]
# # testdata_Level <- testdata_Level[["difficult"]]
#
#
# # ajustes para facilitar o trabalho
# testdata_Level <- testdata_Level#[1:220]

synonym_suggest <- function(text) {
  convert <- our_tokenizer(text)

  # separo os elementos com pontuação
  convert1 <- convert[!grepl(" |\\,|\\.|\\?|\\!|\\%|\\(|\\)|\\]|\\[|\\:|<|>", convert)] # nao remover nem - nem ' nem ’
  pontuacao <- convert[grepl(" |\\,|\\.|\\?|\\!|\\%|\\(|\\)|\\]|\\[|\\:|<|>", convert)]
  posit_pontuacao <- grepl(" |\\,|\\.|\\?|\\!|\\%|\\(|\\)|\\]|\\[|\\:|<|>", convert) #
  posit_NA <- is.na(convert1)
  convert1 <- convert1[!is.na(convert1)]



  convert2 <- convert1
  vazio <- c()
  print("Checando sinônimos")
  for(i in 1:length(convert1)) {
    x <- convert1[[i]]
    n <- grep(paste0("^", x, "$"), Syn_organized$B.word_dif)
    if(is_empty(n)) {
      vazio <- c(vazio, i)
      w <- x
      } else {
      z <- paste(Syn_organized$Syn_easy[n], collapse = " / ")
      w <- paste(x, " (", z, ")", sep = "")
      convert2[i] <- w
    }
  }
  # rm("w", "x", "z", "n", "i")
  # convert2

  # print("Checando sinônimos dos lemmas")
  #vazio1 <- c()
  for(i in vazio) {
    x <- our_lemmatizer2(convert2[i], lemmasDF.2.1, return = "first")[[1]]
    n <- grep(paste0("^",x, "$"), Syn_organized$B.word_dif)
    if(is_empty(n)) {
      #vazio1 <- c(vazio1, i)
      w <- x
    } else {
      z <- paste(Syn_organized$Syn_easy[n], collapse = " / ")
      w <- paste(convert2[i], " (", z, ")", sep = "")
      convert2[[i]] <- w
    }
  }
  # rm("w", "x", "z", "n", "i")


  # print("recriando a pontuação")
  final <- rep(NA, length(convert))
  p <- 1
  c <- 1
  for(i in 1:length(final)) {
    if(posit_pontuacao[i] == T) {
      final[i] <- pontuacao[p]
      p <- p+1
    } else {
      final[i] <- convert2[c]
      c <- c+1
    }
  }
  # rm("p", "c", "i")


  # Legendas_selec
  # paste(convert, collapse = "", sep = "")
  # convert
  sinonimos <- paste(final, collapse = "", sep = "")
  return(sinonimos)
}


############## ----









##################################
### (2) Criando função que executa as sugestões de Antonimos


# example <- "https://www.youtube.com/watch?v=4GsjEPRMzdw" # easy
# example <- "https://www.youtube.com/watch?v=LsFdROZ2OdA" # difficult
# Legendas <- youtubecaption::get_caption(url = example,
#                                         language = "en",  # ATENCAO PARA A ESCOLHA DA LEGENDA
#                                         savexl = FALSE, openxl = FALSE, path = getwd())
#
# Legendas_selec <- Legendas$text[20:60]
# LegendasU <- paste(Legendas_selec, collapse= " ")
# LegendasU
#
# LL <- our_tokenizer(LegendasU)
# LL
#
# testdata_Level <- LL
#
# # setwd("C:/Users/Paren/Dropbox/Learning_Community/LCtestcontentR/data")
# # load(file="testdata_Level.RData")
# # load(file="Antony.1.0.Rdata")
# # testdata_Level <- testdata_Level[["easy"]]
# # testdata_Level <- testdata_Level[["difficult"]]
#
#
# # ajustes para facilitar o trabalho
# testdata_Level <- testdata_Level
# convert <- testdata_Level
#

antonym_suggest <- function(text) {
  convert <- our_tokenizer(text)
  # separo os elementos com pontuação
  convert1 <- convert[!grepl(" |\\,|\\.|\\?|\\!|\\%|\\(|\\)|\\]|\\[|\\:|<|>", convert)] # nao remover nem - nem ' nem ’
  pontuacao <- convert[grepl(" |\\,|\\.|\\?|\\!|\\%|\\(|\\)|\\]|\\[|\\:|<|>", convert)]
  posit_pontuacao <- grepl(" |\\,|\\.|\\?|\\!|\\%|\\(|\\)|\\]|\\[|\\:|<|>", convert) #
  posit_NA <- is.na(convert1)
  convert1 <- convert1[!is.na(convert1)]


  convert2 <- convert1
  vazio <- c()
  print("Checando antônimos")
  for(i in 1:length(convert1)) {
    x <- convert1[[i]]
    n <- grep(paste0("^", x, "$"), Ant_organized$B.word_dif)
    if(is_empty(n)) {
      vazio <- c(vazio, i)
      w <- x
      } else {
      z <- paste(Ant_organized$Ant_easy[n], collapse = " / ")
      w <- paste(x, " (", z, ")", sep = "")
      convert2[i] <- w
    }
  }
  # rm("w", "x", "z", "n", "i")
  # convert2

  # print("Checando antônimos dos lemmas")
  #vazio1 <- c()
  for(i in vazio) {
  x <- our_lemmatizer2(convert2[i], lemmasDF.2.2, return = "first")[[1]]
    n <- grep(paste0("^",x, "$"), Ant_organized$B.word_dif)
    if(is_empty(n)) {
      #vazio1 <- c(vazio1, i)
      w <- x
    } else {
      z <- paste(Ant_organized$Ant_easy[n], collapse = " / ")
      w <- paste(convert2[i], " (", z, ")", sep = "")
      convert2[[i]] <- w
    }
  }
  # rm("w", "x", "z", "n", "i")


  # print("recriando a pontuação")
  final <- rep(NA, length(convert))
  p <- 1
  c <- 1
  for(i in 1:length(final)) {
    if(posit_pontuacao[i] == T) {
      final[i] <- pontuacao[p]
      p <- p+1
    } else {
      final[i] <- convert2[c]
      c <- c+1
    }
  }
  # rm("p", "c", "i")


  # Legendas_selec
  # paste(convert, collapse = "", sep = "")
  # convert
  antonimos <- paste(final, collapse = "", sep = "")
  return(antonimos)
}