Porter Stemming Errors

A number of stemming errors have been mentioned in academic papers,  a few are listed below:

 

The Porter stemmer groups “general”, “generous”, “generation”, and “generic” under the same stem, while terms like “recognize” and “recognition” would not be grouped [Hull, 1996 Ref: W6 ]

 

Overstemming errors [Yamout 2004 Ref W7]

 

appendicitis
======
append
=====

paste
=====

past

=====

dying
=====

died

=====
political

=====

politic

=====

polite

=====

comb

=====

combative

=====

general

=====

generative

=====

wit

=====

witness

=====

admire

=====

admiral

=====

engine

=====

engineer

=====

response
=====

responsible

=====

several

=====

severance

=====

experiment
=====

experience
=====

secret
=====

secretion

=====

design
=====

designate

=====

colon
=====

colonize

=====

positive
=====

position

=====

amen
=====

amenity

=====

universe

=====

university

=====