R function for detecting language in text

langScore.R

# Program name: langScore.R
# Author: Alberto Negron
# date: 15/01/2013
# Description: Detect language in plain text

langScore <- function(post) 
{
  # Params:
  # post: Plain text.
  require(tm)
  # Languages supported by tm package
  langList <- list('danish', 'dutch', 'english', 'finnish', 'french', 'german', 'hungarian',
                   'italian', 'norwegian', 'portuguese', 'russian', 'spanish','swedish')
  # List for storing stopwords.
  langStopwords  <- list()
  for (lang in langList) langStopwords[[lang]]  <-  stopwords(kind=lang)
  # Result list
  results <-list() 
  # convert text to list
  toList <- strsplit(post, " ")
  # Normalize tokens
  lowerList <- lapply(toList,tolower)
  
  for(word in lowerList) 
  {
    for(lang in names(langStopwords)) {
      results[[lang]] <- is.element(word, strsplit(langStopwords[[lang]]," "))
    }
    
  }
  scores <- as.vector(lapply(results,sum))
  # In case language is not supported or anything else like links or hashtags
  if(Reduce("+",scores)==0) { return('unknown')}
  max_score <- which.max(scores)
  return(names(scores[max_score]))
}