PCQRSCANER/venv/Lib/site-packages/nltk/test/paice.doctest

36 lines
1.2 KiB
Plaintext
Raw Normal View History

2019-12-22 21:51:47 +01:00
=====================================================
PAICE's evaluation statistics for stemming algorithms
=====================================================
Given a list of words with their real lemmas and stems according to stemming algorithm under evaluation,
counts Understemming Index (UI), Overstemming Index (OI), Stemming Weight (SW) and Error-rate relative to truncation (ERRT).
>>> from nltk.metrics import Paice
-------------------------------------
Understemming and Overstemming values
-------------------------------------
>>> lemmas = {'kneel': ['kneel', 'knelt'],
... 'range': ['range', 'ranged'],
... 'ring': ['ring', 'rang', 'rung']}
>>> stems = {'kneel': ['kneel'],
... 'knelt': ['knelt'],
... 'rang': ['rang', 'range', 'ranged'],
... 'ring': ['ring'],
... 'rung': ['rung']}
>>> p = Paice(lemmas, stems)
>>> p.gumt, p.gdmt, p.gwmt, p.gdnt
(4.0, 5.0, 2.0, 16.0)
>>> p.ui, p.oi, p.sw
(0.8..., 0.125..., 0.15625...)
>>> p.errt
1.0
>>> [('{0:.3f}'.format(a), '{0:.3f}'.format(b)) for a, b in p.coords]
[('0.000', '1.000'), ('0.000', '0.375'), ('0.600', '0.125'), ('0.800', '0.125')]