Sir I am working on classification project before that i have to do feature selection process. i am very interested to apply lemmatization rather stemming. So i executed the code below according to definition of lemmatization it should give root words like run for running and ran, think for thought etc... but my code is not giving correct output.. could you please Please help me to finding out where actually i went wrong please sir.. CODE: library("tm") library("NLP") library("wordnet") setDict("C:/Program Files/WordNet/2.1/dict") vector.documents <- c("The children something to the playground The cars %s down the avenue") corpus.documents <- Corpus(VectorSource(vector.documents)) initDict("C:/Program Files/WordNet/2.1/dict") lapply(corpus.documents,function(x){ sapply(unlist(strsplit(as.character(x),"[[:space:]]+")), function(word) { x.filter <- getTermFilter("StartsWithFilter", word, TRUE) x.filter x terms <- getIndexTerms("NOUN",1,x.filter) terms if(!is.null(terms)) sapply(terms,getLemma) }) }) OUTPUT: $`1` $`1`$The [1] "the absurd" $`1`$children NULL $`1`$playing [1] "playing" $`1`$playground [1] "playground" $`1`$The [1] "the absurd" $`1`$cars [1] "carson" $`1`$landing [1] "landing" $`1`$avenue [1] "avenue" I also tried by applying other POS and type like "Containsfilter" but that also not worked please please help me !!! Thanks in advance. with regards, BHANUMATHI H M [[alternative HTML version deleted]]