By giving some collections of words, the application checks for errors and returns the concatenated collection called "repairedCollection.txt" Using the above file, it produces the "stemmedWords.txt", containing all the words after having stemmed.
By giving the ideal collection and the collection produced by the stemmer, the application ensures to return the evaluation metric that have been selected.
There are some indicative collections at ./evalCollection/ of the stemmer distribution. You are encouraged to use ./test collections/wordCollectionRepository.txt as ideal collection and ./test collections/stemmedWords.txt as stemmed collection.