Tom Brøndsted: Stemming

Stemmer - Upload | Back to Document Classifier

Version 1.1 (Classical Porter stemmer for English ported to PHP 4. Other languges are not supported anymore, since this server does'nt allow execution of "native code")

Input text for stemming

Select stemmer (language):

A stemmer is an algorithm or "device" that brings derived forms of words back to their "stem", e.g. walked back to walk, walks back to walk, walking back to walk etc. This online demo takes an English text as input and attempts to convert each of the words to its stem. Stemming is used for instance in some Information Retrieval Systems similar to Google. Stemming needs not be 100 % "linguistically correct" in order to improve the precision of a search system! Sydney may become Sydnei! No problem as long as there is no other city "Sydnei" that can be confused with "Sydney"! You may try to copy this Intro into the text field and submit it!