Module: Hobix::Stemmable

Defined in:
lib/hobix/search/porter_stemmer.rb

Constant Summary collapse

STEMMED = 
{}
STEP_2_LIST = 
{
  'ational'=>'ate', 'tional'=>'tion', 'enci'=>'ence', 'anci'=>'ance',
  'izer'=>'ize', 'bli'=>'ble',
  'alli'=>'al', 'entli'=>'ent', 'eli'=>'e', 'ousli'=>'ous',
  'ization'=>'ize', 'ation'=>'ate',
  'ator'=>'ate', 'alism'=>'al', 'iveness'=>'ive', 'fulness'=>'ful',
  'ousness'=>'ous', 'aliti'=>'al',
  'iviti'=>'ive', 'biliti'=>'ble', 'logi'=>'log'
}
STEP_3_LIST = 
{
  'icate'=>'ic', 'ative'=>'', 'alize'=>'al', 'iciti'=>'ic',
  'ical'=>'ic', 'ful'=>'', 'ness'=>''
}
SUFFIX_1_REGEXP = 
/(
ational  |
tional   |
enci     |
anci     |
izer     |
bli      |
alli     |
entli    |
eli      |
ousli    |
ization  |
ation    |
ator     |
alism    |
iveness  |
fulness  |
ousness  |
aliti    |
iviti    |
biliti   |
logi)$/x
SUFFIX_2_REGEXP = 
/(
al       |
ance     |
ence     |
er       |
ic       | 
able     |
ible     |
ant      |
ement    |
ment     |
ent      |
ou       |
ism      |
ate      |
iti      |
ous      |
ive      |
ize)$/x
C =

consonant

"[^aeiou]"
V =

vowel

"[aeiouy]"
CC =

consonant sequence

"#{C}(?>[^aeiouy]*)"
VV =

vowel sequence

"#{V}(?>[aeiou]*)"
MGR0 =

[cc]vvcc… is m>0

/^(#{CC})?#{VV}#{CC}/o
MEQ1 =
cc]vvcc[vv

is m=1

/^(#{CC})?#{VV}#{CC}(#{VV})?$/o
MGR1 =

[cc]vvccvvcc… is m>1

/^(#{CC})?#{VV}#{CC}#{VV}#{CC}/o
VOWEL_IN_STEM =

vowel in stem

/^(#{CC})?#{V}/o

Class Method Summary collapse

Class Method Details

.stem_porter(w = self.to_str.dup) ⇒ Object

Porter stemmer in Ruby.

This is the Porter stemming algorithm, ported to Ruby from the version coded up in Perl. It’s easy to follow against the rules in the original paper in:

Porter, 1980, An algorithm for suffix stripping, Program, Vol. 14,
no. 3, pp 130-137,

See also www.tartarus.org/~martin/PorterStemmer

Send comments to [email protected]



101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
 
# File 'lib/hobix/search/porter_stemmer.rb', line 101

def stem_porter(w = self.to_str.dup)

  # make a copy of the given object and convert it to a string.
  original_word = w
  
  return w if w.length < 3

  result = STEMMED[w]
  return result if result

  # now map initial y to Y so that the patterns never treat it as vowel
  w[0] = 'Y' if w[0] == ?y
  
  # Step 1a
  if w =~ /(ss|i)es$/
    w = $` + $1
  elsif w =~ /([^s])s$/ 
    w = $` + $1
  end

  # Step 1b
  if w =~ /eed$/
    w.chop! if $` =~ MGR0 
  elsif w =~ /(ed|ing)$/
    stem = $`
    if stem =~ VOWEL_IN_STEM 
      w = stem
	case w
      when /(at|bl|iz)$/             then w << "e"
      when /([^aeiouylsz])\1$/       then w.chop!
      when /^#{CC}#{V}[^aeiouwxy]$/o then w << "e"
      end
    end
  end

  if w =~ /y$/ 
    stem = $`
    w = stem + "i" if stem =~ VOWEL_IN_STEM 
  end

  # Step 2
  if w =~ SUFFIX_1_REGEXP
    stem = $`
    suffix = $1
    # print "stem= " + stem + "\n" + "suffix=" + suffix + "\n"
    if stem =~ MGR0
      w = stem + STEP_2_LIST[suffix]
    end
  end

  # Step 3
  if w =~ /(icate|ative|alize|iciti|ical|ful|ness)$/
    stem = $`
    suffix = $1
    if stem =~ MGR0
      w = stem + STEP_3_LIST[suffix]
    end
  end

  # Step 4
  if w =~ SUFFIX_2_REGEXP
    stem = $`
    if stem =~ MGR1
      w = stem
    end
  elsif w =~ /(s|t)(ion)$/
    stem = $` + $1
    if stem =~ MGR1
      w = stem
    end
  end

  #  Step 5
  if w =~ /e$/ 
    stem = $`
    if (stem =~ MGR1) ||
        (stem =~ MEQ1 && stem !~ /^#{CC}#{V}[^aeiouwxy]$/o)
      w = stem
    end
  end

  if w =~ /ll$/ && w =~ MGR1
    w.chop!
  end

  # and turn initial Y back to y
  w[0] = 'y' if w[0] == ?Y

  STEMMED[original_word] = w

  w
end