rbbt-text 1.3.4 → 1.3.7
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/LICENSE +20 -0
- data/lib/rbbt/document/annotation.rb +2 -2
- data/lib/rbbt/document/corpus/pubmed.rb +14 -5
- data/lib/rbbt/document/corpus.rb +10 -7
- data/lib/rbbt/document.rb +7 -3
- data/lib/rbbt/ner/abner.rb +3 -2
- data/lib/rbbt/ner/banner.rb +3 -1
- data/lib/rbbt/ner/brat.rb +1 -1
- data/lib/rbbt/ner/linnaeus.rb +2 -1
- data/lib/rbbt/ner/oscar3.rb +0 -1
- data/lib/rbbt/ner/oscar4.rb +0 -1
- data/lib/rbbt/ner/rner.rb +229 -0
- data/lib/rbbt/ner/rnorm/tokens.rb +3 -1
- data/lib/rbbt/ner/rnorm.rb +5 -1
- data/lib/rbbt/ner/token_trieNER.rb +2 -1
- data/lib/rbbt/nlp/open_nlp/sentence_splitter.rb +1 -1
- data/lib/rbbt/nlp/spaCy.rb +158 -15
- data/lib/rbbt/relationship.rb +24 -0
- data/lib/rbbt/segment/named_entity.rb +4 -0
- data/lib/rbbt/segment/range_index.rb +1 -1
- data/lib/rbbt/segment/transformed.rb +9 -1
- data/lib/rbbt/segment.rb +3 -0
- data/share/install/software/OpenNLP +3 -8
- data/share/rner/config.rb +51 -0
- data/test/rbbt/document/corpus/test_pubmed.rb +1 -1
- data/test/rbbt/document/test_annotation.rb +10 -1
- data/test/rbbt/document/test_corpus.rb +14 -0
- data/test/rbbt/ner/rnorm/test_tokens.rb +11 -0
- data/test/rbbt/ner/test_rner.rb +132 -0
- data/test/rbbt/ner/test_rnorm.rb +5 -0
- data/test/rbbt/segment/test_named_entity.rb +2 -1
- data/test/rbbt/segment/test_transformed.rb +13 -30
- data/test/test_spaCy.rb +113 -1
- metadata +13 -18
data/lib/rbbt/nlp/spaCy.rb
CHANGED
@@ -2,30 +2,55 @@ require 'rbbt/segment'
|
|
2
2
|
require 'rbbt/document'
|
3
3
|
require 'rbbt/segment/annotation'
|
4
4
|
require 'rbbt/util/python'
|
5
|
+
require 'rbbt/network/paths'
|
5
6
|
|
6
7
|
module SpaCy
|
7
8
|
|
8
|
-
|
9
|
+
TOKEN_PROPERTIES = %w(lemma_ is_punct is_space shape_ pos_ tag_)
|
10
|
+
CHUNK_PROPERTIES = %w(lemma_)
|
9
11
|
|
10
|
-
def self.
|
12
|
+
def self.nlp(lang = 'en_core_web_md')
|
13
|
+
@@nlp ||= {}
|
14
|
+
@@nlp[lang] ||= RbbtPython.run :spacy do
|
15
|
+
spacy.load(lang)
|
16
|
+
end
|
17
|
+
end
|
18
|
+
|
19
|
+
def self.tokens(text, lang = 'en_core_web_sm')
|
11
20
|
|
12
21
|
tokens = []
|
13
|
-
|
14
|
-
|
15
|
-
|
16
|
-
|
17
|
-
|
18
|
-
|
22
|
+
|
23
|
+
nlp = nlp(lang)
|
24
|
+
doc = nlp.call(text)
|
25
|
+
|
26
|
+
doc.__len__.times do |i|
|
27
|
+
tokens << doc.__getitem__(i)
|
28
|
+
end
|
29
|
+
|
30
|
+
tokens
|
31
|
+
end
|
32
|
+
|
33
|
+
def self.chunks(text, lang = 'en_core_web_sm')
|
34
|
+
|
35
|
+
tokens = []
|
36
|
+
nlp = nlp(lang)
|
37
|
+
|
38
|
+
doc = nlp.call(text)
|
39
|
+
chunks = doc.noun_chunks.__iter__
|
40
|
+
|
41
|
+
RbbtPython.iterate chunks do |item|
|
42
|
+
tokens << item
|
19
43
|
end
|
44
|
+
|
20
45
|
tokens
|
21
46
|
end
|
22
47
|
|
23
|
-
def self.segments(text, lang = '
|
24
|
-
docid = text.docid if Document === text
|
48
|
+
def self.segments(text, lang = 'en_core_web_sm')
|
49
|
+
docid = text.docid if Document === text
|
25
50
|
corpus = text.corpus if Document === text
|
26
51
|
tokens = self.tokens(text, lang).collect do |token|
|
27
52
|
info = {}
|
28
|
-
|
53
|
+
TOKEN_PROPERTIES.each do |p|
|
29
54
|
info[p] = token.instance_eval(p.to_s)
|
30
55
|
end
|
31
56
|
info[:type] = "SpaCy"
|
@@ -35,7 +60,120 @@ module SpaCy
|
|
35
60
|
info[:corpus] = corpus if corpus
|
36
61
|
SpaCyToken.setup(token.text, info)
|
37
62
|
end
|
38
|
-
|
63
|
+
|
64
|
+
tokens
|
65
|
+
end
|
66
|
+
|
67
|
+
def self.chunk_segments(text, lang = 'en_core_web_sm')
|
68
|
+
docid = text.docid if Document === text
|
69
|
+
corpus = text.corpus if Document === text
|
70
|
+
chunks = self.chunks(text, lang).collect do |chunk|
|
71
|
+
info = {}
|
72
|
+
CHUNK_PROPERTIES.each do |p|
|
73
|
+
info[p] = chunk.instance_eval(p.to_s)
|
74
|
+
end
|
75
|
+
start = eend = nil
|
76
|
+
deps = []
|
77
|
+
RbbtPython.iterate chunk.__iter__ do |token|
|
78
|
+
start = token.idx if start.nil?
|
79
|
+
eend = start + chunk.text.length if eend.nil?
|
80
|
+
deps << token.idx.to_s + ":" + token.dep_ + "->" + token.head.idx.to_s if token.head.idx < start || token.head.idx > eend
|
81
|
+
end
|
82
|
+
info[:type] = "SpaCy"
|
83
|
+
info[:offset] = chunk.__iter__.__next__.idx
|
84
|
+
info[:dep] = deps * ";"
|
85
|
+
info[:docid] = docid if docid
|
86
|
+
info[:corpus] = corpus if corpus
|
87
|
+
SpaCySpan.setup(chunk.text, info)
|
88
|
+
end
|
89
|
+
|
90
|
+
chunks
|
91
|
+
end
|
92
|
+
|
93
|
+
def self.dep_graph(text, reverse = false, lang = 'en_core_web_md')
|
94
|
+
tokens = self.segments(text, lang)
|
95
|
+
index = Segment.index(tokens)
|
96
|
+
associations = {}
|
97
|
+
tokens.each do |token|
|
98
|
+
type, target_pos = token.dep.split("->")
|
99
|
+
target_tokens = index[target_pos.to_i]
|
100
|
+
associations[token.segid] = target_tokens
|
101
|
+
end
|
102
|
+
|
103
|
+
if reverse
|
104
|
+
old = associations.dup
|
105
|
+
old.each do |s,ts|
|
106
|
+
ts.each do |t|
|
107
|
+
associations[t] ||= []
|
108
|
+
associations[t] += [s] unless associations[t].include?(s)
|
109
|
+
end
|
110
|
+
end
|
111
|
+
end
|
112
|
+
|
113
|
+
associations
|
114
|
+
end
|
115
|
+
|
116
|
+
def self.chunk_dep_graph(text, reverse = false, lang = 'en_core_web_md')
|
117
|
+
associations = dep_graph(text, false, lang)
|
118
|
+
|
119
|
+
chunks = self.chunk_segments(text, lang)
|
120
|
+
tokens = self.segments(text, lang)
|
121
|
+
index = Segment.index(tokens + chunks)
|
122
|
+
|
123
|
+
chunks.each do |chunk|
|
124
|
+
target_token_ids = chunk.dep.split(";").collect do|dep|
|
125
|
+
type, target_pos = dep.split("->")
|
126
|
+
index[target_pos.to_i]
|
127
|
+
end.flatten
|
128
|
+
|
129
|
+
target_tokens = target_token_ids.collect do |target_token_id|
|
130
|
+
range = Range.new(*target_token_id.split(":").last.split("..").map(&:to_i))
|
131
|
+
range.collect do |pos|
|
132
|
+
index[pos]
|
133
|
+
end.uniq
|
134
|
+
end.flatten
|
135
|
+
associations[chunk.segid] = target_tokens
|
136
|
+
end
|
137
|
+
|
138
|
+
if reverse
|
139
|
+
old = associations.dup
|
140
|
+
old.each do |s,ts|
|
141
|
+
ts.each do |t|
|
142
|
+
associations[t] ||= []
|
143
|
+
associations[t] += [s] unless associations[t].include?(s)
|
144
|
+
end
|
145
|
+
end
|
146
|
+
end
|
147
|
+
|
148
|
+
associations
|
149
|
+
end
|
150
|
+
|
151
|
+
def self.paths(text, source, target, reverse = true, lang = 'en_core_web_md')
|
152
|
+
graph = SpaCy.chunk_dep_graph(text, reverse, lang)
|
153
|
+
|
154
|
+
chunk_index = Segment.index(SpaCy.chunk_segments(text, lang))
|
155
|
+
|
156
|
+
source_id = chunk_index[source.offset].first || source.segid
|
157
|
+
target_id = chunk_index[target.offset].first || target.segid
|
158
|
+
|
159
|
+
path = Paths.dijkstra(graph, source_id, [target_id])
|
160
|
+
|
161
|
+
return nil if path.nil?
|
162
|
+
|
163
|
+
path.reverse
|
164
|
+
end
|
165
|
+
|
166
|
+
def self.config(base, target = nil)
|
167
|
+
TmpFile.with_file(base) do |baseconfig|
|
168
|
+
if target
|
169
|
+
CMD.cmd(:spacy, "init fill-config #{baseconfig} #{target}")
|
170
|
+
else
|
171
|
+
TmpFile.with_file do |tmptarget|
|
172
|
+
CMD.cmd(:spacy, "init fill-config #{baseconfig} #{tmptarget}")
|
173
|
+
Open.read(targetconfig)
|
174
|
+
end
|
175
|
+
end
|
176
|
+
end
|
39
177
|
end
|
40
178
|
end
|
41
179
|
|
@@ -43,10 +181,15 @@ module SpaCyToken
|
|
43
181
|
extend Entity
|
44
182
|
include SegmentAnnotation
|
45
183
|
|
46
|
-
self.annotation *SpaCy::
|
184
|
+
self.annotation *SpaCy::TOKEN_PROPERTIES
|
47
185
|
self.annotation :dep
|
48
186
|
end
|
49
187
|
|
50
|
-
|
51
|
-
|
188
|
+
module SpaCySpan
|
189
|
+
extend Entity
|
190
|
+
include SegmentAnnotation
|
191
|
+
|
192
|
+
self.annotation *SpaCy::CHUNK_PROPERTIES
|
193
|
+
self.annotation :dep
|
52
194
|
end
|
195
|
+
|
@@ -0,0 +1,24 @@
|
|
1
|
+
require 'rbbt/segment'
|
2
|
+
|
3
|
+
module Relationship
|
4
|
+
extend Annotation
|
5
|
+
self.annotation :segment
|
6
|
+
self.annotation :terms
|
7
|
+
self.annotation :type
|
8
|
+
|
9
|
+
def text
|
10
|
+
if segment
|
11
|
+
segment
|
12
|
+
else
|
13
|
+
type + ": " + terms * ", "
|
14
|
+
end
|
15
|
+
end
|
16
|
+
|
17
|
+
def html
|
18
|
+
text = <<-EOF
|
19
|
+
<span class='Relationship'\
|
20
|
+
>#{ self.text }</span>
|
21
|
+
EOF
|
22
|
+
text.chomp
|
23
|
+
end
|
24
|
+
end
|
@@ -6,7 +6,7 @@ module Segment::RangeIndex
|
|
6
6
|
SegID.setup(res, :corpus => corpus)
|
7
7
|
end
|
8
8
|
|
9
|
-
def self.index(segments, corpus, persist_file = :memory)
|
9
|
+
def self.index(segments, corpus = nil, persist_file = :memory)
|
10
10
|
segments = segments.values.flatten if Hash === segments
|
11
11
|
|
12
12
|
annotation_index =
|
@@ -70,7 +70,15 @@ module Transformed
|
|
70
70
|
orig_length = self.length
|
71
71
|
|
72
72
|
offset = self.respond_to?(:offset) ? self.offset.to_i : 0
|
73
|
-
|
73
|
+
|
74
|
+
segments = segments.select do |s|
|
75
|
+
shift = shift s.range
|
76
|
+
s_offset = s.offset.to_i
|
77
|
+
s_offset += shift.first if shift
|
78
|
+
|
79
|
+
s_offset >= offset &&
|
80
|
+
s_offset <= offset + self.length - 1
|
81
|
+
end
|
74
82
|
|
75
83
|
Segment.clean_sort(segments).each do |segment|
|
76
84
|
next if segment.offset.nil?
|
data/lib/rbbt/segment.rb
CHANGED
@@ -1,12 +1,7 @@
|
|
1
1
|
#!/bin/bash
|
2
2
|
|
3
3
|
name='OpenNLP'
|
4
|
-
url="http://apache.rediris.es/opennlp/opennlp-1.9.
|
4
|
+
url="http://apache.rediris.es/opennlp/opennlp-1.9.4/apache-opennlp-1.9.4-bin.tar.gz"
|
5
5
|
|
6
|
-
|
7
|
-
|
8
|
-
|
9
|
-
|
10
|
-
ln -sf "$OPT_DIR/$name/lib/"*.jar "$OPT_JAR_DIR/"
|
11
|
-
|
12
|
-
clean_build
|
6
|
+
install_src $name $url
|
7
|
+
(cd $OPT_DIR/jars; ln -s $OPT_DIR/$name/lib/*.jar .)
|
@@ -0,0 +1,51 @@
|
|
1
|
+
isLetters /^[A-Z]+$/i
|
2
|
+
isUpper /^[A-Z]+$/
|
3
|
+
isLower /^[a-z]+$/
|
4
|
+
isDigits /^[0-9]+$/i
|
5
|
+
isRoman /^[IVX]+$/
|
6
|
+
isGreek /^(?:alpha|beta|gamma|delta|epsilon|zeta|eta|theta|iota|kappa|lambda|mu|nu|xi|omicron|pi|rho|sigma|tau|upsilon|phi|chi|psi|omega)$/i
|
7
|
+
isPunctuation /^[,.;]$/
|
8
|
+
isDelim /^[\/()\[\]{}\-]$/
|
9
|
+
isNonWord /^[^\w]+$/
|
10
|
+
isConjunction /^and|or|&|,$/
|
11
|
+
|
12
|
+
hasLetters /[A-Z]/i
|
13
|
+
hasUpper /.[A-Z]/
|
14
|
+
hasLower /[a-z]/
|
15
|
+
hasDigits /[0-9]/i
|
16
|
+
hasGreek /(?:alpha|beta|gamma|delta|epsilon|zeta|eta|theta|iota|kappa|lambda|mu|nu|xi|omicron|pi|rho|sigma|tau|upsilon|phi|chi|psi|omega)/i
|
17
|
+
hasPunctuation /[,.;]/
|
18
|
+
hasDelim /[\/()\[\]{}\-]/
|
19
|
+
hasNonWord /[^\w]/
|
20
|
+
caspMix /[a-z].[A-Z]/
|
21
|
+
keywords /(?:protein|gene|domain|ase)s?$/
|
22
|
+
hasSuffix /[a-z][A-Z0-9]$/
|
23
|
+
|
24
|
+
numLetters do |w| w.scan(/[A-Z]/i).length end
|
25
|
+
numDigits do |w| w.scan(/[0-9]/).length end
|
26
|
+
#
|
27
|
+
prefix_3 /^(...)/
|
28
|
+
prefix_4 /^(....)/
|
29
|
+
suffix_3 /(...)$/
|
30
|
+
suffix_4 /(....)$/
|
31
|
+
|
32
|
+
|
33
|
+
token1 do |w|
|
34
|
+
w.sub(/[A-Z]/,'A').
|
35
|
+
sub(/[a-z]/,'a').
|
36
|
+
sub(/[0-9]/,'0').
|
37
|
+
sub(/[^0-9a-z]/i,'x')
|
38
|
+
end
|
39
|
+
token2 do |w|
|
40
|
+
w.sub(/[A-Z]+/,'A').
|
41
|
+
sub(/[a-z]+/,'a').
|
42
|
+
sub(/[0-9]+/,'0').
|
43
|
+
sub(/[^0-9a-z]+/i,'x')
|
44
|
+
end
|
45
|
+
token3 do |w| w.downcase end
|
46
|
+
special do |w| w.is_special? end
|
47
|
+
|
48
|
+
context %w(special token2 isPunctuation isDelim)
|
49
|
+
window %w(1 2 3 -1 -2 -3)
|
50
|
+
#direction :reverse
|
51
|
+
|
@@ -7,7 +7,7 @@ class TestCorpusPubmed < Test::Unit::TestCase
|
|
7
7
|
def test_add_pmid
|
8
8
|
corpus = Document::Corpus.setup({})
|
9
9
|
|
10
|
-
document = corpus.add_pmid("
|
10
|
+
document = corpus.add_pmid("33359141", :abstract).first
|
11
11
|
title = document.to(:title)
|
12
12
|
assert title.include?("COVID-19")
|
13
13
|
end
|
@@ -4,6 +4,7 @@ require 'rbbt/document/corpus'
|
|
4
4
|
require 'rbbt/segment'
|
5
5
|
require 'rbbt/document/annotation'
|
6
6
|
require 'rbbt/segment/named_entity'
|
7
|
+
require 'rbbt/ner/abner'
|
7
8
|
|
8
9
|
class TestAnnotation < Test::Unit::TestCase
|
9
10
|
class CalledOnce < Exception; end
|
@@ -28,6 +29,12 @@ class TestAnnotation < Test::Unit::TestCase
|
|
28
29
|
self.split(" ").collect{|e| NamedEntity.setup(e, :code => Misc.digest(e)) }
|
29
30
|
end
|
30
31
|
|
32
|
+
Document.define :abner do
|
33
|
+
$called_once = true
|
34
|
+
Abner.new.match(self)
|
35
|
+
end
|
36
|
+
|
37
|
+
|
31
38
|
Document.persist :ner
|
32
39
|
end
|
33
40
|
|
@@ -133,7 +140,9 @@ class TestAnnotation < Test::Unit::TestCase
|
|
133
140
|
text.ner
|
134
141
|
|
135
142
|
assert ! $called_once
|
136
|
-
|
143
|
+
|
144
|
+
assert_equal text.abner.first.docid, text.docid
|
145
|
+
|
137
146
|
assert text.ner.first.segid.include?("TEST:")
|
138
147
|
end
|
139
148
|
end
|
@@ -29,5 +29,19 @@ class TestDocumentCorpus < Test::Unit::TestCase
|
|
29
29
|
assert corpus.docids("TEST:").include?(text.docid)
|
30
30
|
end
|
31
31
|
end
|
32
|
+
|
33
|
+
def test_load
|
34
|
+
text = "This is a document"
|
35
|
+
Document.setup(text, "TEST", "test_doc1", nil)
|
36
|
+
|
37
|
+
TmpFile.with_file do |path|
|
38
|
+
corpus = Persist.open_tokyocabinet(path, true, :single, "BDB")
|
39
|
+
corpus.extend Document::Corpus
|
40
|
+
|
41
|
+
corpus.add_document(text)
|
42
|
+
|
43
|
+
assert corpus.docids("TEST:").include?(text.docid)
|
44
|
+
end
|
45
|
+
end
|
32
46
|
end
|
33
47
|
|
@@ -0,0 +1,11 @@
|
|
1
|
+
require File.join(File.expand_path(File.dirname(__FILE__)), '../../..', 'test_helper.rb')
|
2
|
+
require 'rbbt/ner/rnorm'
|
3
|
+
|
4
|
+
class TestRNorm < Test::Unit::TestCase
|
5
|
+
def test_evaluate
|
6
|
+
t = Tokenizer.new
|
7
|
+
assert t.evaluate("PDGFRA","PDGFRalpha") > 0
|
8
|
+
iii t.evaluate("JUNB","JunB")
|
9
|
+
end
|
10
|
+
end
|
11
|
+
|
@@ -0,0 +1,132 @@
|
|
1
|
+
require File.dirname(__FILE__) + '/../../test_helper'
|
2
|
+
require 'rbbt'
|
3
|
+
require 'rbbt/ner/rner'
|
4
|
+
require 'test/unit'
|
5
|
+
|
6
|
+
class TestRNer < Test::Unit::TestCase
|
7
|
+
|
8
|
+
def setup
|
9
|
+
@parser = NERFeatures.new() do
|
10
|
+
isLetters /^[A-Z]+$/i
|
11
|
+
context prefix_3 /^(...)/
|
12
|
+
downcase do |w| w.downcase end
|
13
|
+
|
14
|
+
context %w(downcase)
|
15
|
+
end
|
16
|
+
end
|
17
|
+
|
18
|
+
def test_config
|
19
|
+
config = <<-EOC
|
20
|
+
isLetters /^[A-Z]+$/i
|
21
|
+
context prefix_3 /^(...)/
|
22
|
+
downcase do |w| w.downcase end
|
23
|
+
|
24
|
+
context %w(downcase)
|
25
|
+
EOC
|
26
|
+
|
27
|
+
assert_equal config.strip, @parser.config.strip
|
28
|
+
end
|
29
|
+
|
30
|
+
def test_reverse
|
31
|
+
assert_equal("protein P53", NERFeatures.reverse("P53 protein"))
|
32
|
+
assert_equal(
|
33
|
+
". LH of assay - radioimmuno serum the with compared was LH urinary for ) GONAVIS - HI ( test hemagglutination direct new A",
|
34
|
+
NERFeatures.reverse(
|
35
|
+
"A new direct hemagglutination test (HI-GONAVIS) for urinary LH was compared with the serum\n radioimmuno-assay of LH."
|
36
|
+
))
|
37
|
+
end
|
38
|
+
|
39
|
+
def test_features
|
40
|
+
assert_equal @parser.features("abCdE"), ["abCdE",true,'abC','abcde']
|
41
|
+
end
|
42
|
+
|
43
|
+
def test_template
|
44
|
+
template =<<-EOT
|
45
|
+
UisLetters: %x[0,1]
|
46
|
+
Uprefix_3: %x[0,2]
|
47
|
+
Uprefix_3#1: %x[1,2]
|
48
|
+
Uprefix_3#-1: %x[-1,2]
|
49
|
+
Udowncase: %x[0,3]
|
50
|
+
Udowncase#1: %x[1,3]
|
51
|
+
Udowncase#-1: %x[-1,3]
|
52
|
+
B
|
53
|
+
EOT
|
54
|
+
|
55
|
+
assert(@parser.template == template)
|
56
|
+
end
|
57
|
+
|
58
|
+
def test_tokens
|
59
|
+
assert( NERFeatures.tokens("A new direct hemagglutination test (HI-GONAVIS) for urinary LH was compared with the serum\n radioimmuno-assay of LH.")==
|
60
|
+
["A", "new", "direct", "hemagglutination", "test", "(", "HI", "-", "GONAVIS", ")", "for", "urinary", "LH", "was", "compared", "with", "the", "serum", "radioimmuno", "-", "assay", "of", "LH", "."])
|
61
|
+
|
62
|
+
|
63
|
+
end
|
64
|
+
def test_text_features
|
65
|
+
|
66
|
+
assert(@parser.text_features("abCdE 1234") == [["abCdE",true, "abC", "abcde"], ["1234",false, "123", "1234"]])
|
67
|
+
assert(@parser.text_features("abCdE 1234",true) == [["abCdE",true, "abC", "abcde",1], ["1234",false, "123", "1234",2]])
|
68
|
+
assert(@parser.text_features("abCdE 1234",false) == [["abCdE",true, "abC", "abcde",0], ["1234",false, "123", "1234",0]])
|
69
|
+
|
70
|
+
end
|
71
|
+
|
72
|
+
def test_tagged_features
|
73
|
+
assert_equal(
|
74
|
+
[["phosphorilation",true, "pho", "phosphorilation", 0],
|
75
|
+
["of",true, false, "of", 0],
|
76
|
+
["GENE1",false, "GEN", "gene1", 1],
|
77
|
+
[".", false, false, ".", 0]],
|
78
|
+
@parser.tagged_features("phosphorilation of GENE1.",['GENE1']))
|
79
|
+
|
80
|
+
assert_equal(
|
81
|
+
[["GENE1",false, "GEN", "gene1", 1],
|
82
|
+
["phosphorilation",true, "pho", "phosphorilation", 0]],
|
83
|
+
@parser.tagged_features("GENE1 phosphorilation",['GENE1']))
|
84
|
+
|
85
|
+
|
86
|
+
assert_equal(
|
87
|
+
[["phosphorilation",true, "pho", "phosphorilation", 0],
|
88
|
+
["of",true, false, "of", 0],
|
89
|
+
["GENE",true, "GEN", "gene", 1],
|
90
|
+
["1",false, false, "1", 2],
|
91
|
+
[".", false, false, ".", 0]],
|
92
|
+
@parser.tagged_features("phosphorilation of GENE 1.",['GENE 1']))
|
93
|
+
end
|
94
|
+
|
95
|
+
def test_tagged_features_reverse
|
96
|
+
@parser.reverse = true
|
97
|
+
assert_equal(
|
98
|
+
[
|
99
|
+
["GENE1",false, "GEN", "gene1", 1],
|
100
|
+
["of",true, false, "of", 0],
|
101
|
+
["phosphorilation",true, "pho", "phosphorilation", 0]
|
102
|
+
],
|
103
|
+
@parser.tagged_features("phosphorilation of GENE1",['GENE1']))
|
104
|
+
|
105
|
+
assert_equal(
|
106
|
+
[
|
107
|
+
[".", false, false, ".", 0],
|
108
|
+
["1",false, false, "1", 1],
|
109
|
+
["GENE",true, "GEN", "gene", 2],
|
110
|
+
["of",true, false, "of", 0],
|
111
|
+
["phosphorilation",true, "pho", "phosphorilation", 0]
|
112
|
+
],
|
113
|
+
@parser.tagged_features("phosphorilation of GENE 1.",['GENE 1']))
|
114
|
+
end
|
115
|
+
|
116
|
+
def test_default_config
|
117
|
+
require 'rbbt/bow/misc'
|
118
|
+
text =<<-EOF
|
119
|
+
This text explains how MDM2 interacts with TP53.
|
120
|
+
EOF
|
121
|
+
@parser = NERFeatures.new Rbbt.share.rner["config.rb"].find
|
122
|
+
features = @parser.tagged_features text, %w(TP53 MDM2)
|
123
|
+
assert features.first.first == "This"
|
124
|
+
end
|
125
|
+
|
126
|
+
|
127
|
+
|
128
|
+
def __test_CRFPP_install
|
129
|
+
assert(require File.join(Rbbt.datadir, 'third_party/crf++/ruby/CRFPP'))
|
130
|
+
end
|
131
|
+
|
132
|
+
end
|
data/test/rbbt/ner/test_rnorm.rb
CHANGED
@@ -24,7 +24,8 @@ class TestClass < Test::Unit::TestCase
|
|
24
24
|
|
25
25
|
def test_tsv
|
26
26
|
a = "test"
|
27
|
-
NamedEntity.setup a, 10, "TYPE", "CODE", "SCORE"
|
27
|
+
NamedEntity.setup a, 10, "DocID", "TYPE", "CODE", "SCORE"
|
28
|
+
ppp Annotated.tsv([a,a])
|
28
29
|
assert Annotated.tsv([a]).fields.include? "code"
|
29
30
|
assert Annotated.tsv([a], nil).fields.include? "code"
|
30
31
|
assert Annotated.tsv([a], :all).fields.include? "code"
|
@@ -144,7 +144,7 @@ More recently, PPAR activators were shown to inhibit the activation of inflammat
|
|
144
144
|
gene2.entity_type = "Protein"
|
145
145
|
|
146
146
|
Transformed.with_transform(a, [gene1,gene2], Proc.new{|e| e.html}) do
|
147
|
-
assert_equal "This sentence mentions the <span class='Entity' attr-entity-type='Gene'>TP53</span> gene and the <span class='Entity' attr-entity-type='Protein'>CDK5R1</span> protein", a
|
147
|
+
assert_equal "This sentence mentions the <span class='Entity' attr-entity-type='Gene' title='Gene'>TP53</span> gene and the <span class='Entity' attr-entity-type='Protein' title='Protein'>CDK5R1</span> protein", a
|
148
148
|
end
|
149
149
|
end
|
150
150
|
|
@@ -165,7 +165,7 @@ More recently, PPAR activators were shown to inhibit the activation of inflammat
|
|
165
165
|
gene2.entity_type = "Protein"
|
166
166
|
|
167
167
|
Transformed.with_transform(a, [gene1,gene2], Proc.new{|e| e.html}) do
|
168
|
-
assert_equal "This sentence mentions the <span class='Entity' attr-entity-type='Gene'>TP53</span> gene and the <span class='Entity' attr-entity-type='Protein'>CDK5R1</span> protein", a
|
168
|
+
assert_equal "This sentence mentions the <span class='Entity' attr-entity-type='Gene' title='Gene'>TP53</span> gene and the <span class='Entity' attr-entity-type='Protein' title='Protein'>CDK5R1</span> protein", a
|
169
169
|
end
|
170
170
|
end
|
171
171
|
|
@@ -185,9 +185,9 @@ More recently, PPAR activators were shown to inhibit the activation of inflammat
|
|
185
185
|
assert_equal [gene1], Segment.overlaps(Segment.sort([gene1,gene2]))
|
186
186
|
|
187
187
|
Transformed.with_transform(a, [gene1], Proc.new{|e| e.html}) do
|
188
|
-
assert_equal "This sentence mentions the <span class='Entity' attr-entity-type='Gene'>TP53</span> gene and the CDK5R1 protein", a
|
188
|
+
assert_equal "This sentence mentions the <span class='Entity' attr-entity-type='Gene' title='Gene'>TP53</span> gene and the CDK5R1 protein", a
|
189
189
|
Transformed.with_transform(a, [gene2], Proc.new{|e| e.html}) do
|
190
|
-
assert_equal "This sentence mentions the <span class='Entity' attr-entity-type='Expanded Gene'><span class='Entity' attr-entity-type='Gene'>TP53</span> gene</span> and the CDK5R1 protein", a
|
190
|
+
assert_equal "This sentence mentions the <span class='Entity' attr-entity-type='Expanded Gene' title='Expanded Gene'><span class='Entity' attr-entity-type='Gene' title='Gene'>TP53</span> gene</span> and the CDK5R1 protein", a
|
191
191
|
end
|
192
192
|
end
|
193
193
|
end
|
@@ -393,43 +393,26 @@ This is another sentence. Among the nonstructural proteins, the leader protein (
|
|
393
393
|
end
|
394
394
|
end
|
395
395
|
|
396
|
-
def
|
397
|
-
a = "
|
396
|
+
def test_transform_sorter_end
|
397
|
+
a = "The transcription factors farnesoid X receptor, small heterodimer partner, liver receptor homolog-1, and liver X receptor comprise the signaling cascade network that regulates the expression and secretion of apoM."
|
398
398
|
original = a.dup
|
399
399
|
|
400
|
-
gene1 = "
|
400
|
+
gene1 = "liver receptor homolog-1"
|
401
401
|
gene1.extend Segment
|
402
402
|
gene1.offset = a.index gene1
|
403
403
|
|
404
|
-
gene2 = "
|
404
|
+
gene2 = "apoM"
|
405
405
|
gene2.extend Segment
|
406
406
|
gene2.offset = a.index gene2
|
407
407
|
|
408
408
|
assert_equal gene1, a[gene1.range]
|
409
409
|
assert_equal gene2, a[gene2.range]
|
410
410
|
|
411
|
-
|
412
|
-
|
413
|
-
|
414
|
-
|
415
|
-
|
416
|
-
assert_equal c, Transformed.transform(a,[gene1], "GN")
|
417
|
-
|
418
|
-
iii a.transformation_offset_differences
|
419
|
-
raise
|
420
|
-
assert_equal gene2.offset, a.transformation_offset_differences.first.first.first
|
421
|
-
assert_equal gene1.offset, a.transformation_offset_differences.last.first.first
|
422
|
-
|
423
|
-
|
424
|
-
gene3 = "GN gene"
|
425
|
-
gene3.extend Segment
|
426
|
-
gene3.offset = a.index gene3
|
427
|
-
|
428
|
-
assert_equal gene3, a[gene3.range]
|
429
|
-
|
430
|
-
a.restore([gene3])
|
431
|
-
assert_equal original, a
|
432
|
-
assert_equal "TP53 gene", a[gene3.range]
|
411
|
+
Transformed.with_transform(a, [gene1], "[TF]") do
|
412
|
+
Transformed.with_transform(a, [gene2], "[TG]") do
|
413
|
+
assert_equal "The transcription factors farnesoid X receptor, small heterodimer partner, [TF], and liver X receptor comprise the signaling cascade network that regulates the expression and secretion of [TG].", a
|
414
|
+
end
|
415
|
+
end
|
433
416
|
|
434
417
|
end
|
435
418
|
|