browser_web_data_entity_sumarization 1.0.0beta1
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/lib/browser_web_data_entity_sumarization.rb +23 -0
- data/lib/browser_web_data_entity_sumarization/entity_sumarization_nif_parser.rb +59 -0
- data/lib/browser_web_data_entity_sumarization/entity_sumarization_predicate.rb +33 -0
- data/lib/browser_web_data_entity_sumarization/entity_sumarization_predicates_similarity.rb +263 -0
- data/lib/browser_web_data_entity_sumarization/entity_sumarization_statistics.rb +609 -0
- data/lib/browser_web_data_entity_sumarization/sparql_request.rb +74 -0
- data/lib/browser_web_data_entity_sumarization/version.rb +3 -0
- data/lib/config/entity_sumarization_config.rb +48 -0
- data/lib/knowledge/classes_hierarchy.json +906 -0
- data/lib/knowledge/common_properties.json +23 -0
- data/lib/knowledge/entity_classes.json +1 -0
- data/lib/knowledge/knowledge_base.json +40642 -0
- data/lib/utils/cache_helper.rb +69 -0
- data/lib/utils/hash_helper.rb +79 -0
- data/lib/utils/sparql_queries.rb +126 -0
- data/lib/utils/string_helper.rb +31 -0
- metadata +74 -0
@@ -0,0 +1,69 @@
|
|
1
|
+
#encoding: utf-8
|
2
|
+
|
3
|
+
# Module CacheHelper
|
4
|
+
module CacheHelper
|
5
|
+
|
6
|
+
###
|
7
|
+
# The method clear all application cache
|
8
|
+
#
|
9
|
+
# @param [String] type File extension type.
|
10
|
+
def self.clear_cache(type = '.json')
|
11
|
+
dir = "#{Dir.tmpdir}/#{BrowserWebData::TMP_DIR}/*#{type}"
|
12
|
+
Dir.glob(dir).each { |path|
|
13
|
+
FileUtils.rm_f(path)
|
14
|
+
}
|
15
|
+
end
|
16
|
+
|
17
|
+
##
|
18
|
+
# The method helps to load cached json.
|
19
|
+
# This cache is permanent and reload only if no exist or set demand_reload
|
20
|
+
#
|
21
|
+
# @param [UU::OS::UESURI] binary_ap_uri UU::OS::UESURI of property to load with caching.
|
22
|
+
# @param [Hash] params Load parameters
|
23
|
+
# @option params [Hash] :json Optional parameters. Json parse attributes. Default is {symbolize_names:true}.
|
24
|
+
# @option params [Fixnum] :ttl Optional parameters. Time to live in second, for this time duration will be load property from json_cache file. Default is 10800.
|
25
|
+
# @option params [Boolean] :demanded_reload Optional parameters. Flag to reload value from property. Default is false.
|
26
|
+
#
|
27
|
+
# @return [Hash] hash_value
|
28
|
+
#
|
29
|
+
# @yield return value must be Hash
|
30
|
+
def self.load_cached(key, params = {}, &block)
|
31
|
+
default_load_attrs = {
|
32
|
+
update: false,
|
33
|
+
json: {symbolize_names: true},
|
34
|
+
ttl: 0,
|
35
|
+
demanded_reload: false
|
36
|
+
}
|
37
|
+
params = default_load_attrs.merge(params)
|
38
|
+
hash = {}
|
39
|
+
|
40
|
+
cache_dir_path = "#{Dir.tmpdir}/#{BrowserWebData::TMP_DIR}"
|
41
|
+
Dir.mkdir(cache_dir_path) unless Dir.exist?(cache_dir_path)
|
42
|
+
cache_file_path = "#{cache_dir_path}/#{StringHelper.get_clear_file_path(key)}.json"
|
43
|
+
|
44
|
+
if params[:demanded_reload] || !File.exists?(cache_file_path) || (params[:ttl] && Time.now - File.ctime(cache_file_path) > params[:ttl])
|
45
|
+
|
46
|
+
if block_given?
|
47
|
+
hash = yield hash
|
48
|
+
File.open(cache_file_path, 'w') { |f| f.puts hash.to_json } unless hash.empty?
|
49
|
+
end
|
50
|
+
else
|
51
|
+
hash = JSON.parse(File.read(cache_file_path).force_encoding('UTF-8'), params[:json])
|
52
|
+
|
53
|
+
end
|
54
|
+
|
55
|
+
HashHelper.recursive_symbolize_keys(hash)
|
56
|
+
end
|
57
|
+
|
58
|
+
###
|
59
|
+
# The method helps to get build in knowledge by key.
|
60
|
+
#
|
61
|
+
# @param [String] key
|
62
|
+
def self.load_knowledge(key)
|
63
|
+
dir_path = "#{File.dirname(File.expand_path('..', __FILE__))}/knowledge"
|
64
|
+
file_path = "#{dir_path}/#{StringHelper.get_clear_file_path(key)}.json"
|
65
|
+
|
66
|
+
JSON.parse(File.read(file_path), symbolize_names: true)
|
67
|
+
end
|
68
|
+
|
69
|
+
end
|
@@ -0,0 +1,79 @@
|
|
1
|
+
#encoding: utf-8
|
2
|
+
|
3
|
+
|
4
|
+
# Module HashHelper
|
5
|
+
module HashHelper
|
6
|
+
|
7
|
+
##
|
8
|
+
# The method helps to get new sorted hash by key.
|
9
|
+
#
|
10
|
+
# @param [Hash] hash Input hash which will be sorted.
|
11
|
+
# @param [Symbol, String] type Type of sorting, default is asc as ascending. One of [:asc, :desc]
|
12
|
+
#
|
13
|
+
# @return [Hash] sorted_hash
|
14
|
+
def self.get_sorted(hash, type = :asc)
|
15
|
+
hash = {} unless hash
|
16
|
+
case type.to_s.downcase.to_sym
|
17
|
+
when :asc
|
18
|
+
Hash[hash.sort]
|
19
|
+
when :desc
|
20
|
+
Hash[hash.sort{|a,b| a<=>b}]
|
21
|
+
else
|
22
|
+
hash
|
23
|
+
end
|
24
|
+
end
|
25
|
+
|
26
|
+
##
|
27
|
+
# The method recursively symbolizes keys of hash.
|
28
|
+
#
|
29
|
+
# @param [Hash, Enumerable] input_value Data to by symbolized.
|
30
|
+
# @return [Hash, Enumerable] Symbolized data.
|
31
|
+
def self.recursive_symbolize_keys(input_value)
|
32
|
+
case input_value
|
33
|
+
when Hash
|
34
|
+
Hash[
|
35
|
+
input_value.map do |k, v|
|
36
|
+
[k.respond_to?(:to_sym) ? k.to_sym : k, recursive_symbolize_keys(v)]
|
37
|
+
end
|
38
|
+
]
|
39
|
+
when Enumerable
|
40
|
+
input_value.map { |v| recursive_symbolize_keys(v) }
|
41
|
+
else
|
42
|
+
input_value
|
43
|
+
end
|
44
|
+
end
|
45
|
+
|
46
|
+
##
|
47
|
+
# The method recursively unsymbolizes keys of hash.
|
48
|
+
#
|
49
|
+
# @param [Hash, Enumerable] input_value Data to by symbolized.
|
50
|
+
# @return [Hash, Enumerable] Symbolized data.
|
51
|
+
def self.recursive_unsymbolize_keys(input_value)
|
52
|
+
case input_value
|
53
|
+
when Hash
|
54
|
+
Hash[
|
55
|
+
input_value.map do |k, v|
|
56
|
+
[k.respond_to?(:to_s) ? k.to_s : k, recursive_unsymbolize_keys(v)]
|
57
|
+
end
|
58
|
+
]
|
59
|
+
when Enumerable
|
60
|
+
input_value.map { |v| recursive_unsymbolize_keys(v) }
|
61
|
+
else
|
62
|
+
input_value
|
63
|
+
end
|
64
|
+
end
|
65
|
+
|
66
|
+
|
67
|
+
def self.recursive_map_keys(data)
|
68
|
+
data.map{|k,v|
|
69
|
+
if v.is_a?(Hash) && !v.empty?
|
70
|
+
inner_array = recursive_map_keys(v)
|
71
|
+
else
|
72
|
+
inner_array = []
|
73
|
+
end
|
74
|
+
|
75
|
+
[k] + inner_array
|
76
|
+
}.reduce(:+)
|
77
|
+
end
|
78
|
+
|
79
|
+
end
|
@@ -0,0 +1,126 @@
|
|
1
|
+
# encoding: utf-8
|
2
|
+
|
3
|
+
module SPARQLQueries
|
4
|
+
|
5
|
+
def resources_by_dbpedia_page_rank(entity_type, limit = 10)
|
6
|
+
entity_type = entity_type['http'] ? "<#{entity_type}>" : "dbo:#{entity_type}"
|
7
|
+
|
8
|
+
" PREFIX rdf:<http://www.w3.org/1999/02/22-rdf-syntax-ns#>
|
9
|
+
PREFIX dbo:<http://dbpedia.org/ontology/>
|
10
|
+
PREFIX vrank:<http://purl.org/voc/vrank#>
|
11
|
+
|
12
|
+
SELECT ?entity ?rank
|
13
|
+
FROM <http://dbpedia.org>
|
14
|
+
FROM <http://people.aifb.kit.edu/ath/#DBpedia_PageRank>
|
15
|
+
WHERE {
|
16
|
+
?entity rdf:type #{entity_type}.
|
17
|
+
?entity vrank:hasRank/vrank:rankValue ?rank.
|
18
|
+
}
|
19
|
+
ORDER BY DESC(?rank) LIMIT #{limit}"
|
20
|
+
end
|
21
|
+
|
22
|
+
def all_predicates_by_object(object)
|
23
|
+
object = object['http'] ? "<#{object}>" : "dbo:#{object}"
|
24
|
+
|
25
|
+
" PREFIX dbo: <http://dbpedia.org/ontology/>
|
26
|
+
PREFIX dbp: <http://dbpedia.org/property/>
|
27
|
+
|
28
|
+
SELECT DISTINCT ?property
|
29
|
+
|
30
|
+
WHERE {
|
31
|
+
?subject ?property #{object}.
|
32
|
+
}"
|
33
|
+
end
|
34
|
+
|
35
|
+
def all_predicates_by_subject(subject, only_literal)
|
36
|
+
subject = subject['http'] ? "<#{subject}>" : "dbo:#{subject}"
|
37
|
+
filter = only_literal ? 'FILTER(isLiteral(?object))' : nil
|
38
|
+
|
39
|
+
" PREFIX dbo: <http://dbpedia.org/ontology/>
|
40
|
+
PREFIX dbp: <http://dbpedia.org/property/>
|
41
|
+
|
42
|
+
SELECT DISTINCT ?property
|
43
|
+
|
44
|
+
WHERE {
|
45
|
+
#{subject} ?property ?object.
|
46
|
+
#{filter}
|
47
|
+
}"
|
48
|
+
end
|
49
|
+
|
50
|
+
def all_predicates_by_object_and_subject(subject, object)
|
51
|
+
subject = subject['http'] ? "<#{subject}>" : "dbo:#{subject}"
|
52
|
+
object = object['http'] ? "<#{object}>" : "dbo:#{object}"
|
53
|
+
|
54
|
+
" PREFIX dbo: <http://dbpedia.org/ontology/>
|
55
|
+
PREFIX dbp: <http://dbpedia.org/property/>
|
56
|
+
|
57
|
+
SELECT DISTINCT ?property
|
58
|
+
|
59
|
+
WHERE {
|
60
|
+
#{subject} ?property #{object}.
|
61
|
+
}"
|
62
|
+
end
|
63
|
+
|
64
|
+
def count_predicate_by_entity(entity_class, predicate)
|
65
|
+
entity_class = entity_class['http'] ? "<#{entity_class}>" : "dbo:#{entity_class}"
|
66
|
+
predicate = predicate['http'] ? "<#{predicate}>" : "dbo:#{predicate}"
|
67
|
+
|
68
|
+
" PREFIX dbo: <http://dbpedia.org/ontology/>
|
69
|
+
PREFIX dbp: <http://dbpedia.org/property/>
|
70
|
+
|
71
|
+
SELECT DISTINCT COUNT(?subject) as ?count
|
72
|
+
|
73
|
+
WHERE {
|
74
|
+
?subject a #{entity_class} .
|
75
|
+
{?subject #{predicate} ?a .} UNION {?b #{predicate} ?subject .}
|
76
|
+
}
|
77
|
+
|
78
|
+
ORDER BY DESC(?count)"
|
79
|
+
end
|
80
|
+
|
81
|
+
def count_of_identical_predicates(predicates)
|
82
|
+
predicates = [predicates] unless predicates.is_a?(Array)
|
83
|
+
where_part = predicates.map{|predicate|
|
84
|
+
predicate = predicate['http'] ? "<#{predicate}>" : "dbo:#{predicate}"
|
85
|
+
"?subject #{predicate} ?object ."
|
86
|
+
}.join("\n")
|
87
|
+
|
88
|
+
" SELECT COUNT(DISTINCT ?subject) AS ?count
|
89
|
+
WHERE{#{where_part}
|
90
|
+
}"
|
91
|
+
end
|
92
|
+
|
93
|
+
def resource_properties(resource, lang = 'en')
|
94
|
+
resource = resource['http'] ? "<#{resource}>" : "<http://dbpedia.org/resource/#{resource}>"
|
95
|
+
|
96
|
+
" PREFIX dbo: <http://dbpedia.org/ontology/>
|
97
|
+
PREFIX dbp: <http://dbpedia.org/property/>
|
98
|
+
SELECT DISTINCT ?predicate, ?predicate_label, ?value, ?value_label
|
99
|
+
WHERE {
|
100
|
+
{ #{resource} ?predicate ?value . } UNION { ?value ?predicate #{resource} . }
|
101
|
+
|
102
|
+
OPTIONAL{
|
103
|
+
?value rdfs:label ?value_label .
|
104
|
+
FILTER (lang(?value_label) = '#{lang}')
|
105
|
+
}
|
106
|
+
|
107
|
+
?predicate rdfs:label ?predicate_label .
|
108
|
+
FILTER (lang(?predicate_label) = '#{lang}')
|
109
|
+
}"
|
110
|
+
end
|
111
|
+
|
112
|
+
def entity_classes(resource)
|
113
|
+
resource = resource['http'] ? "<#{resource}>" : "<http://dbpedia.org/resource/#{resource}"
|
114
|
+
|
115
|
+
" SELECT DISTINCT ?entity_class
|
116
|
+
WHERE {
|
117
|
+
#{resource} a ?entity_class .
|
118
|
+
?entity_class a owl:Class .
|
119
|
+
}"
|
120
|
+
end
|
121
|
+
|
122
|
+
def self.included(base)
|
123
|
+
base.extend SPARQLQueries
|
124
|
+
end
|
125
|
+
|
126
|
+
end
|
@@ -0,0 +1,31 @@
|
|
1
|
+
#encoding: utf-8
|
2
|
+
|
3
|
+
|
4
|
+
# Module StringHelper
|
5
|
+
module StringHelper
|
6
|
+
|
7
|
+
##
|
8
|
+
# The method helps to replace problematic chars from string to be used as part of file path.
|
9
|
+
#
|
10
|
+
# @param [String] path
|
11
|
+
#
|
12
|
+
# @return [String] path
|
13
|
+
def self.get_clear_file_path(path)
|
14
|
+
path.to_s.gsub(/[:\/\.\*#]/, '_')
|
15
|
+
end
|
16
|
+
|
17
|
+
##
|
18
|
+
# The method helps to get snake case string from camel case one.
|
19
|
+
#
|
20
|
+
# @param [String] path
|
21
|
+
#
|
22
|
+
# @return [String] snake_cased_string
|
23
|
+
def self.get_snake_case(string)
|
24
|
+
string.to_s.gsub(/::/, '/').
|
25
|
+
gsub(/([A-Z]+)([A-Z][a-z])/,'\1_\2').
|
26
|
+
gsub(/([a-z\d])([A-Z])/,'\1_\2').
|
27
|
+
tr('-', '_').
|
28
|
+
downcase
|
29
|
+
end
|
30
|
+
|
31
|
+
end
|
metadata
ADDED
@@ -0,0 +1,74 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: browser_web_data_entity_sumarization
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 1.0.0beta1
|
5
|
+
platform: ruby
|
6
|
+
authors:
|
7
|
+
- Marek Filteš
|
8
|
+
autorequire:
|
9
|
+
bindir: bin
|
10
|
+
cert_chain: []
|
11
|
+
date: 2017-04-12 00:00:00.000000000 Z
|
12
|
+
dependencies:
|
13
|
+
- !ruby/object:Gem::Dependency
|
14
|
+
requirement: !ruby/object:Gem::Requirement
|
15
|
+
requirements:
|
16
|
+
- - '='
|
17
|
+
- !ruby/object:Gem::Version
|
18
|
+
version: 2.1.0
|
19
|
+
name: sparql-client
|
20
|
+
prerelease: false
|
21
|
+
type: :runtime
|
22
|
+
version_requirements: !ruby/object:Gem::Requirement
|
23
|
+
requirements:
|
24
|
+
- - '='
|
25
|
+
- !ruby/object:Gem::Version
|
26
|
+
version: 2.1.0
|
27
|
+
description:
|
28
|
+
email: marek.filtes@gmail.com
|
29
|
+
executables: []
|
30
|
+
extensions: []
|
31
|
+
extra_rdoc_files: []
|
32
|
+
files:
|
33
|
+
- lib/browser_web_data_entity_sumarization.rb
|
34
|
+
- lib/browser_web_data_entity_sumarization/entity_sumarization_nif_parser.rb
|
35
|
+
- lib/browser_web_data_entity_sumarization/entity_sumarization_predicate.rb
|
36
|
+
- lib/browser_web_data_entity_sumarization/entity_sumarization_predicates_similarity.rb
|
37
|
+
- lib/browser_web_data_entity_sumarization/entity_sumarization_statistics.rb
|
38
|
+
- lib/browser_web_data_entity_sumarization/sparql_request.rb
|
39
|
+
- lib/browser_web_data_entity_sumarization/version.rb
|
40
|
+
- lib/config/entity_sumarization_config.rb
|
41
|
+
- lib/knowledge/classes_hierarchy.json
|
42
|
+
- lib/knowledge/common_properties.json
|
43
|
+
- lib/knowledge/entity_classes.json
|
44
|
+
- lib/knowledge/knowledge_base.json
|
45
|
+
- lib/utils/cache_helper.rb
|
46
|
+
- lib/utils/hash_helper.rb
|
47
|
+
- lib/utils/sparql_queries.rb
|
48
|
+
- lib/utils/string_helper.rb
|
49
|
+
homepage:
|
50
|
+
licenses:
|
51
|
+
- MIT
|
52
|
+
metadata: {}
|
53
|
+
post_install_message:
|
54
|
+
rdoc_options: []
|
55
|
+
require_paths:
|
56
|
+
- lib
|
57
|
+
- results
|
58
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
59
|
+
requirements:
|
60
|
+
- - ">="
|
61
|
+
- !ruby/object:Gem::Version
|
62
|
+
version: '0'
|
63
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
64
|
+
requirements:
|
65
|
+
- - ">"
|
66
|
+
- !ruby/object:Gem::Version
|
67
|
+
version: 1.3.1
|
68
|
+
requirements: []
|
69
|
+
rubyforge_project:
|
70
|
+
rubygems_version: 2.4.8
|
71
|
+
signing_key:
|
72
|
+
specification_version: 4
|
73
|
+
summary: Tool for entity sumarization.
|
74
|
+
test_files: []
|